skip to main content


Search for: All records

Creators/Authors contains: "Sun, Hao"

Note: When clicking on a Digital Object Identifier (DOI) number, you will be taken to an external site maintained by the publisher. Some full text articles may not yet be available without a charge during the embargo (administrative interval).
What is a DOI Number?

Some links on this page may take you to non-federal websites. Their policies may differ from this site.

  1. Biomass-derived polymer materials are emerging as sustainable and low-carbon footprint alternatives to the current petroleum-based commodity plastics. In the past decade, the ring-opening metathesis polymerization (ROMP) technique has been widely used for the polymerization of cyclic olefin monomers derived from biorenewable resources, giving rise to a diverse set of biobased polymer materials. However, most synthetic biobased polymers made by ROMP are nondegradable because of their all-carbon backbones. Herein, we present a modular synthetic strategy to acid-degradable poly(enol ether)s via ring-opening metathesis copolymerization of biorenewable oxanorbornenes and 3,4-dihydropyran (DHP). 1H NMR analysis reveals that the percentage of DHP units in the resulting copolymers gradually increases as the feed ratio of DHP to oxanorbornene increases. The composition of the copolymers plays a pivotal role in governing their thermal properties. Thermogravimetric analysis shows that an increasing percentage of DHP results in a decrease in the decomposition temperatures, suggesting that the incorporation of enol ether groups in the polymer backbone reduces the thermal stability of the copolymers. Moreover, a wide range of glass transition temperatures (16–165 °C) can be achieved by tuning the copolymer composition and the oxanorbornene structure. Critically, all of the poly(enol ether)s developed in this study are degradable under mildly acidic conditions. A higher incorporation of DHP in the copolymer leads to enhanced degradability, as evidenced by smaller final degradation products. Altogether, this study provides a facile approach for synthesizing biorenewable and degradable polymer materials with highly tunable thermal properties desired for their potential industrial applications. 
    more » « less
    Free, publicly-accessible full text available February 28, 2025
  2. The current insufficient recycling of commodity polymer waste has resulted in pressing environmental and human health issues in our modern society. In the quest for next-generation polymer materials, chemists have recently shifted their attention to the design of chemically recyclable polymers that can undergo depolymerization to regenerate monomers under mild conditions. During the past decade, ring-closing metathesis reactions have been demonstrated to be a robust approach for the depolymerization of polyolefins, producing low-strain cyclic alkene products which can be repolymerized back to new batches of polymers. In this review, we aim to highlight the recent advances in chemical recycling of polyolefins enabled by ring-closing metathesis depolymerization (RCMD). A library of depolymerizable polyolefins will be covered based on the ring size of their monomers or depolymerization products, including five-membered, six-membered, eight-membered, and macrocyclic rings. Moreover, current limitations, potential applications, and future opportunities of the RCMD approach will be discussed. It is clear from recent research in this field that RCMD represents a powerful strategy towards closed-loop chemical recycling of novel polyolefin materials. 
    more » « less
    Free, publicly-accessible full text available January 4, 2025
  3. Degradable polymers made via ring-opening metathesis polymerization (ROMP) hold tremendous promise as eco-friendly materials. However, most of the ROMP monomers are derived from petroleum resources, which are typically considered less sustainable compared to biomass. Herein, we present a synthetic strategy to degradable polymers by harnessing alternating ROMP of biomass-based cyclic olefin monomers including exo-oxanorbornenes and cyclic enol ethers. A library of well-defined poly(enol ether)s with modular structures, tunable glass transition temperatures, and controlled molecular weights was achieved, demonstrating the versatility of this approach. Most importantly, the resulting copolymers exhibit high degrees of alternation, rendering their backbones fully degradable under acidic conditions. 
    more » « less
    Free, publicly-accessible full text available November 20, 2024
  4. Abstract

    Traditional data-driven deep learning models often struggle with high training costs, error accumulation, and poor generalizability in complex physical processes. Physics-informed deep learning (PiDL) addresses these challenges by incorporating physical principles into the model. Most PiDL approaches regularize training by embedding governing equations into the loss function, yet this depends heavily on extensive hyperparameter tuning to weigh each loss term. To this end, we propose to leverage physics prior knowledge by “baking” the discretized governing equations into the neural network architecture via the connection between the partial differential equations (PDE) operators and network structures, resulting in a PDE-preserved neural network (PPNN). This method, embedding discretized PDEs through convolutional residual networks in a multi-resolution setting, largely improves the generalizability and long-term prediction accuracy, outperforming conventional black-box models. The effectiveness and merit of the proposed methods have been demonstrated across various spatiotemporal dynamical systems governed by spatiotemporal PDEs, including reaction-diffusion, Burgers’, and Navier-Stokes equations.

     
    more » « less
  5. : In order to evaluate urban earthquake resilience, reliable structural modeling is needed. However, detailed modeling of a large number of structures and carrying out time history analyses for sets of ground motions are not practical at an urban scale. Reduced-order surrogate models can expedite numerical simulations while maintaining necessary engineering accuracy. Neural networks have been shown to be a powerful tool for developing surrogate models, which often outperform classical surrogate models in terms of scalability of complex models. Training a reliable deep learning model, however, requires an immense amount of data that contain a rich input-output relationship, which typically cannot be satisfied in practical applications. In this paper, we propose model-informed symbolic neural networks (MiSNN) that can discover the underlying closed-form formulations (differential equations) for a reduced-order surrogate model. The MiSNN will be trained on datasets obtained from dynamic analyses of detailed reinforced concrete special moment frames designed for San Francisco, California, subject to a series of selected ground motions. Training the MiSNN is equivalent to finding the solution to a sparse optimization problem, which is solved by the Adam optimizer. The earthquake ground acceleration and story displacement, velocity, and acceleration time histories will be used to train 1) an integrated SNN, which takes displacement and velocity states and outputs the absolute acceleration response of the structure; and 2) a distributed SNN, which distills the underlying equation of motion for each story. The results show that the MiSNN can reduce computational cost while maintaining high prediction accuracy of building responses. 
    more » « less
    Free, publicly-accessible full text available July 9, 2024
  6. The large population movement during the Spring Festival travel in China can considerably accelerate the spread of epidemics, especially after the relaxation of strict control measures against COVID-19. This study aims to assess the impact of population migration in Spring Festival holiday on epidemic spread under different scenarios. Using inter-city population movement data, we construct the population flow network during the non-holiday time as well as the Spring Festival holiday. We build a large-scale metapopulation model to simulate the epidemic spread among 371 Chinese cities. We analyze the impact of Spring Festival travel on the peak timing and peak magnitude nationally and in each city. Assuming an R0 (basic reproduction number) of 15 and the initial conditions as the reported COVID-19 infections on 17 December 2022, model simulations indicate that the Spring Festival travel can substantially increase the national peak magnitude of infection. The infection peaks arrive at most cities 1–4 days earlier as compared to those of the non-holiday time. While peak infections in certain large cities, such as Beijing and Shanghai, are decreased due to the massive migration of people to smaller cities during the pre-Spring Festival period, peak infections increase significantly in small- or medium-sized cities. For a less transmissible disease (R0 = 5), infection peaks in large cities are delayed until after the Spring Festival. Small- or medium-sized cities may experience a larger infection due to the large-scale population migration from metropolitan areas. The increased disease burden may impose considerable strain on the healthcare systems in these resource-limited areas. For a less transmissible disease, particular attention needs to be paid to outbreaks in large cities when people resume work after holidays. 
    more » « less
    Free, publicly-accessible full text available July 1, 2024
  7. Abstract

    Clustering analysis of sequence data continues to address many applications in engineering design, aided with the rapid growth of machine learning in applied science. This paper presents an unsupervised machine learning algorithm to extract defining characteristics of earthquake ground‐motion spectra, also called latent features, to aid in ground‐motion selection (GMS). In this context, a latent feature is a low‐dimensional machine‐discovered spectral characteristic learned through nonlinear relationships of a neural network autoencoder. Machine discovered latent features can be combined with traditionally defined intensity measures and clustering can be performed to select a representative subgroup from a large ground‐motion suite. The objective of efficient GMS is to choose characteristic records representative of what the structure will probabilistically experience in its lifetime. Three examples are presented to validate this approach, including the use of synthetic and field recorded ground‐motion datasets. The presented deep embedding clustering of ground‐motion spectra has three main advantages: (1) defining characteristics that represent the sparse spectral content of ground motions are discovered efficiently through training of the autoencoder, (2) domain knowledge is incorporated into the machine learning framework with conditional variables in the deep embedding scheme, and (3) the method results in a ground‐motion subgroup that is more representative of the original ground‐motion suite compared to traditional GMS techniques.

     
    more » « less