Lin Wang

Northwestern University, Evanston, IL, United States

Are you Lin Wang?

Claim your profile

Publications (11)20.87 Total impact

  • [Show abstract] [Hide abstract]
    ABSTRACT: Conventional continuum mechanistic models for polymer degradation typically involve thousands of coupled differential-algebraic equations, requiring an efficient solver to solve the complex set of stiff model equations. This can be overcome by formulating the problem in terms of a stochastic simulation procedure, requiring only iterative operations to solve the model. The present work describes the detailed mechanistic modeling of pyrolysis of poly(styrene peroxide) (PSP) using kinetic Monte Carlo (KMC) simulation to predict the product yields and gain a better understanding of the product evolution pathways. The traditionally accepted mechanism of PSP pyrolysis proposed by Mayo and Miller, which involves the key reaction steps of peroxide bond fission, alkoxy radical recombination and disproportionation, and end chain β-scission, was initially tested using the KMC model to predict the peroxide concentration profile and the product yields. This model was only qualitatively able to predict the major products, benzaldehyde and formaldehyde, while the formation of minor products like α-hydroxy acetophenone, phenyl glycol, and phenyl glyoxal was not captured at all. Hence, a new mechanism that also incorporated hydrogen abstraction and β-scission was proposed and implemented in KMC. The final model tracked 949 reactions of 83 species. The rate coefficients for all the reaction steps were based on the existing literature reports, and hence no parameter estimation was done to fit the model against the experimental data. The revised model was quantitatively able to predict all the products of PSP pyrolysis, which was attributed to the stabilization of the alkoxy radicals by hydrogen abstraction, and the subsequent generation of additional alkoxy radicals by β-scission. KMC allowed the dominant pathways for the formation of minor products and dimers to be identified explicitly. Finally, the implications of this study in understanding the effect of trace peroxide bonds on poly(styrene) pyrolysis are outlined.
    Chemical Engineering Science 02/2012; 69(1):456–471. · 2.61 Impact Factor
  • Lin Wang, Linda J. Broadbelt
    Macromolecular Theory and Simulations 03/2011; 20(3):n/a-n/a. · 1.79 Impact Factor
  • Lin Wang, Linda J. Broadbelt
    Macromolecular Theory and Simulations 02/2011; 20(3):191 - 204. · 1.61 Impact Factor
  • Lin Wang, Linda J. Broadbelt
    Macromolecular Theory and Simulations 10/2010; 20(1):54 - 64. · 1.61 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The nitroxide-mediated controlled radical polymerization (NM-CRP) of 4-acetoxystyrene with an alkoxyamine was analyzed by a combined experimental and modeling approach. At low nitroxide concentrations, thermal initiation was significant, and control of the polydispersity was poor, as was observed previously for styrene. A continuum model based on the method of moments was used to regress the parameters for the reversible nitroxide uncoupling/coupling reactions (activation energy of uncoupling), thermal initiation (activation energy of initiation), and termination (frequency factor of recombination). The model was able to capture the molecular weight averages and the polydispersity index as a function of time and the nitroxide concentration qualitatively and quantitatively. Using this mechanistic framework, we developed kinetic Monte Carlo models that allowed the molecular weight distributions to be predicted explicitly in good agreement with experimental data. A comparison of the NM-CRP of 4-acetoxystyrene and styrene is provided to illustrate the effect of the acetoxy substituent. © 2010 Wiley Periodicals, Inc. J Appl Polym Sci, 2010
    Journal of Applied Polymer Science 05/2010; 118(2):740 - 750. · 1.40 Impact Factor
  • Lin Wang, Linda J. Broadbelt
    [Show abstract] [Hide abstract]
    ABSTRACT: The kinetic details of segment formation in living radical polymerization (LRP) were investigated by studying styrene (S)/methyl methacrylate (MMA) copolymerization via nitroxide-mediated controlled radical polymerization (NM-CRP) using kinetic Monte Carlo (KMC) simulations. It was demonstrated that the classic theory describing the distribution of instantaneous segment lengths developed for conventional free radical polymerization (FRP) does not always hold in LRP. Segment formation can be controlled by deactivation through coupling of growing radicals and nitroxide radicals instead of selective preference during propagation due to different reactivity ratios and composition of competing monomers if the transient lifetime is too short. Mathematical equations were proposed for determining in a facile way whether the segment growth is controlled by deactivation under given reaction conditions. The uniformity of segment growth was analyzed as well. It takes a surprisingly long time to allow the majority of the chains to be reactivated at least once in an S-rich reaction system, which leads to high polydispersity in monomer sequences formed on different chains. It was also demonstrated that there inevitably exists a distribution of segment lengths at a given monomer composition no matter how “ideal” the livingness is. This distribution is determined by the “intrinsic” kinetics of copolymerization and is not manipulated by the features of LRP.
    Macromolecules 02/2010; 43(5). · 5.93 Impact Factor
  • Lin Wang, Linda J. Broadbelt
    [Show abstract] [Hide abstract]
    ABSTRACT: Copolymers with well-defined architectures are increasingly desired in a wide range of applications as the backbone microstructure can have a dramatic impact on the properties of the polymer. For example, gradient copolymers, in which there is a gradual change in composition along the chain length, are a new class of polymeric materials that show superior blending capabilities in polymer compatibilization. Although the monomer-by-monomer sequence along the copolymer chain plays an important role in the research and potential applications of gradient copolymers, few efforts have been directed at the investigation of the explicit sequence of gradient copolymers. Here, we report the development of a simulation framework based on kinetic Monte Carlo written in house, which can predict the explicit sequence formed along each chain by tracking the growth of each individual chain instead of concentration. The framework is generally applicable to various reaction types and can incorporate differences in reactivity of different chain ends in a facile way. We use this framework to investigate the explicit sequence along the backbone of MMA/ S gradient copolymers prepared by NM-CRP in a semi-batch reactor as an example. This system serves as a good example because S/MMA copolymerization by NM-CRP has been widely studied and detailed kinetic parameters are available in literature. KMC simulations are a powerful tool to predict the explicit sequence of copolymers and can serve as a companion to experimental efforts to precisely design the sequence length along copolymer chains. Current efforts to utilize KMC simulations to develop synthesis recipes to meet design targets of sequence patterns along copolymer chains are underway.
    2009 AIChE Annual Meeting; 11/2009
  • Lin Wang, Linda J. Broadbelt
    [Show abstract] [Hide abstract]
    ABSTRACT: Gradient copolymers, which are expected to have an intermediate chain architecture between conventional block copolymers and random copolymers, have attracted much interest in a wide range of applications. Because of the well-ordered variation in chain structure, gradient copolymers have been theorized to exhibit interfacial activities superior to block copolymers of the same composition. However, the sequences in gradient copolymers cannot be determined directly by experimental methods. To address this challenge, we developed kinetic Monte Carlo (KMC) models to predict explicit monomer-by-monomer sequence along the copolymer chain, which is not possible with traditional moment-based continuum models. We demonstrate our approach using styrene/methyl methacrylate gradient copolymerization with nitroxide-mediated controlled radical polymerization(NM-CRP). It was found that the variation in the average sequence length as a function of chain length does not resemble that of the instantaneous composition, which has been considered to be sufficient to characterize gradient copolymers. Our findings indicate that copolymers with compositional gradients may have chain sequences resembling those of statistical copolymers and suggest that characterization of structural gradients is warranted. In order to unravel the impact of monomer sequences on the physical properties of gradient copolymers, there must be a shift from a focus on control of composition to control of sequence. We show that KMC simulations are a powerful tool to predict the explicit sequence of copolymers and can serve as a companion to experimental efforts to precisely design the sequence length distribution of copolymers.
    2009 AIChE Annual Meeting; 11/2009
  • Lin Wang, Linda J. Broadbelt
    [Show abstract] [Hide abstract]
    ABSTRACT: The impact of the synthesis conditions on the formation of the monomer sequences of styrene (S) and methyl methacrylate (MMA) gradient copolymers synthesized by forced gradient copolymerization with nitroxide-mediated controlled radical polymerization (NM-CRP) was investigated using kinetic Monte Carlo (KMC) simulations. The factors affecting the formation of the individual segments, arrangement of these segments along the chain, and the uniformity of monomer sequences were investigated. It was shown that instantaneous segment lengths increase exponentially as a function of monomer composition. The concentration of nitroxyl radicals also plays a key role in the formation of segment lengths. In addition, the arrangement of the segments mainly depends on the feed profile of the second monomer. A constant feed profile, which is widely used in current syntheses of gradient copolymers, is shown to not be suitable to make a structural gradient along the chain. It was also demonstrated that the uniformity of sequence patterns can be affected by the entire growth history of the chains, and thus strong control over the uniformity of chain growth is required throughout the reaction in order to achieve sequences in different chains that resemble one another.
    Macromolecules 11/2009; · 5.93 Impact Factor
  • Lin Wang, Linda J. Broadbelt
    [Show abstract] [Hide abstract]
    ABSTRACT: In spite of many efforts made to study gradient copolymers, the monomer-by-monomer sequence along the chain is still obscure. A general computational framework based on kinetic Monte Carlo simulations was developed to predict the explicit sequence of copolymers. We demonstrate our approach using styrene (S)/methyl methacrylate (MMA) gradient copolymers synthesized by a semibatch process with nitroxide-mediated controlled radical polymerization (NM-CRP). It was found that the variation in the average segment length as a function of chain length does not resemble that of the instantaneous composition. Our findings indicate that copolymers with compositional gradients may have monomer-by-monomer sequences resembling those of statistical copolymers. It was also found that the explicit sequence can significantly deviate from that of the instantaneous composition as a function of chain length when combination is the favored termination mode. These details of explicit sequence revealed by KMC simulations are obscured by only considering fractional composition measures to characterize the gradient shape.
    Macromolecules 10/2009; 42(20). · 5.93 Impact Factor
  • Lin Wang, Linda J. Broadbelt
    [Show abstract] [Hide abstract]
    ABSTRACT: A gradient copolymer has a gradient in the repeat units comprising the backbone of the polymer arranged from predominantly monomer A to predominantly monomer B along the copolymer chain. Because of the gradual change of the composition along the copolymer chains, gradient copolymers exhibit distinct physical properties compared to those of random or block copolymers of the same composition. Furthermore, different sequence distributions along gradient copolymer chains may lead to markedly different physical properties, suggesting that gradient polymers can be tailored for specific applications by tuning the sequence distribution. Thus, it would be valuable to be able to predict the specific sequence distribution of gradient copolymers given different monomer pairs and synthesis conditions. To address this challenge, we developed kinetic Monte Carlo (KMC) models, which track molecules instead of concentration, in order to track the explicit sequence distribution for each copolymer chain. In addition, KMC models permit the explicit molecular weight distribution (MWD) and chemical composition distribution (CCD) of gradient copolymers to be tracked, which is not possible with moment-based continuum models. We developed these KMC models in the context of the nitroxide-mediated controlled radical polymerization (NM-CRP) of two different monomer systems: styrene/4-acetoxystyrene and methyl methacrylate/t-butyl methacrylate. The simulated MWD and CCD were compared to experimental data from matrix-assisted laser desorption/ionization-time of flight mass spectroscopy (MALDI-TOF-MS). The effects of different synthesis conditions on the MWD, CCD and formation of the compositional gradient along the copolymer chains were studied. Finally, experimental methods for obtaining information about the sequence distribution to compare to the detailed KMC output were explored.
    2008 AIChE Annual Meeting; 11/2008