Full Length Article
Computational investigation of hydrogen-induced phonon changes in carbon fiber

https://doi.org/10.1016/j.commatsci.2022.111884Get rights and content

Abstract

Optical vibrational spectroscopy has shown promise as a noninvasive means of monitoring the mechanical properties of carbon fiber (CF), which is increasingly used for industrial and consumer purposes. However, interpretation of optical vibrational spectra for solid materials is inferential, particularly when defects are present. Because inelastic neutron scattering (INS) spectroscopy is not subject to selection rules, the full vibrational spectra can be measured. And, identifying correlations between INS features and tensile properties can assist in the interpretation of spectra from more commonly used optical vibrational spectroscopic techniques, such as Raman and infrared (IR) spectroscopy. Recent INS experiments on high-performance commercial carbon fibers showed features near 900 and 1100 cm−1 in addition to a broad feature near 3000 cm−1 that increased in intensity with decreasing tensile strength. These features were assigned to hydrogen defects. In the present work, we use density functional theory to simulate the INS spectra of several hydrogen defect geometries in graphite as a model for carbon fiber structure units, confirming the experimental assignment of these peaks to hydrogen modes and providing insights into the structure and lattice dynamics of the defects.

Introduction

Carbon fiber (CF) is used in a variety of applications that require the advantages of lightweight and strong materials, such as the aeronautical and automotive industries. However, the mechanical properties of CF vary significantly under different manufacturing conditions [1], [2], [3]. Polyacrylonitrile, or PAN, is the most widely used CF precursor and generally yields fibers with high–tensile strength. It is often combined with acidic comonomers to increase drawability and promote polymerization of the fiber [4]. Mesopitch-based fibers are also produced and usually have a high modulus [5]. The manufacturing steps with the largest influence on the final material strength is the carbonization and graphitization temperatures, which are the temperatures to which the CF is heated to complete the cyclization of the precursor material, remove noncarbon components from the fiber matrix and increase the graphitic orientation of the fiber core. The tensile strength typically increases with increasing treatment temperatures up to 1600 °C; however, fibers are often heated up to 3000 °C to increase fiber modulus at the cost of reduced tensile strength [5]. Many credit the high-temperature increase of the tensile modulus to a decrease in defect density at higher treatment temperatures [5], [6], but this does not explain the abrupt reduction in tensile strength above 1600 °C. As such, the exact relationship between CF processing conditions, fiber microstructure, and tensile properties remains an open question.

To prevent the critical failure of CF during use, it is valuable to monitor the properties of CF materials using a noninvasive technique. Optical vibrational spectroscopy has become the primary tool for such investigations, and Raman spectroscopy in particular has been used extensively to study the vibrational properties of carbonaceous materials due to the perceived simplicity of the spectra [7], [8], [9], [10], [11], [12], [13], [14], [15], [16]. The most prominent peaks are the G peak near 1580 cm−1, the D peak around 1350 cm−1, and the D' peak around 1620 cm−1, though additional peaks have been reported [17], [18]. Despite the presence of only a few peaks, Raman spectroscopy is subject to selection rules, and interpretation of the spectra can become complex, especially when selection rules begin to break down because of defects [19], [20].

As Brubaker et al. showed, extracted information from a Raman spectrum, such as peak locations and widths, is extremely sensitive to the fitting procedure used. Their recommended fitting procedure varies slightly for different fiber modulus classes [20]. Inelastic neutron scattering (INS) experiments are a valuable supplement to Raman spectroscopy because INS has no selection rules. Because of this, Brubaker et al. performed INS experiments to probe the entire vibrational spectra of several commercially available CF [21]. Owing to the large incoherent neutron scattering cross section of hydrogen compared to other prevalent elements in CF, Brubaker et al. identified a correlation between tensile modulus and hydrogen content. A peak around 3000 cm−1 was assigned to carbon–hydrogen stretching modes, and the area under this peak increased (i.e., higher hydrogen concentrations) for CF with lower tensile modulus. Additionally, Brubaker et al. identified INS features around 900–950 cm−1 and 1100 cm−1 that varied with tensile modulus. Guided by previous work on activated carbons, they assigned these features to out-of-plane and in-plane carbon–hydrogen vibrations, respectively [22], [23], [24].

Although the work of Brubaker et al. established a connection between hydrogen defects and the mechanical properties of CF, the atomic structure of the hydrogen defects in CF is not clear. The work on activated carbons investigated hydrogen atoms bound to the edges of graphene flakes, a local environment which does not include the interlayer interactions expected in CF. Previous work has shown that the extent of out-of-plane displacement of hydrogen-bound carbon atoms is reduced by approximately 0.2 Å in graphite compared to graphene because of the interlayer van der Waals interactions between the graphitic layers [25]. Additionally, the carbon–hydrogen bond of hydrogen adsorbed to graphene vacancies, which is highly favorable at standard atmospheric conditions, has been calculated to be 0.01–0.08 Å less than hydrogen adsorbed to pristine graphene [25], [26]. Consequently, it is important to extend the knowledge of the vibrational spectra of hydrogen adsorbed on graphene flakes to include interlayer interactions and hydrogen adsorption to other common CF defects, particularly monovacancies, which we investigated recently [27].

In this work, we have performed density functional theory (DFT) calculations to simulate the INS spectra of several hydrogen defects in graphite as a surrogate for CF, and we have investigated the changes to the phonons induced by these defect structures. In addition to providing support for the experimental peak assignments, we showed that the INS features previously reported for terminal hydrogen adsorption on graphene flakes can be extended to hydrogen substitution and intercalation in bulk graphite, although the designation of in-plane versus out-of-plane bending modes is complicated in the multilayer material. This is especially true when defect–defect interactions are present, which can also notably change the shape and position of the phonon density of states and INS peaks.

Section snippets

Methods

We investigated the vibrational properties of hydrogen-defective graphite using spin-polarized DFT as implemented in the electronic structure theory package Real-space Multigrid (RMG) DFT [28], [29], which was executed on the Summit supercomputer at the Oak Ridge Leadership Computing Facility. RMG uses real-space descriptions of the wave functions, charge densities, potentials, and the Kohn–Sham Hamiltonian to reduce the number of Fourier transforms between real and reciprocal space needed

Defect structures

There are two nonequivalent carbon atoms in pristine graphite. The first, which we refer to as A-type atoms, sit directly between carbon atoms in the neighboring graphene sheets, while the second (B-type carbon atoms) sit directly between hexagon centers (Fig. 1A). As such, there are two nonequivalent substitution sites in graphite, though their formation energies differ by only 0.006 eV. We find the substitutions of B-type carbon atoms are slightly lower in energy (formation energy of 6.03 eV)

Conclusions

In conclusion, we used DFT to simulate the INS spectra of hydrogen substitution and intercalation in bulk graphite to aid in the interpretation of recent experiments on commercially available CF. We provided additional support for the assignment of the peaks near 950, 1100, and 3000 cm−1 to carbon–hydrogen vibrations and showed that the interlayer interactions of graphite are not strong enough to alter the frequency of the modes as reported for activated carbons. However, the designation of

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This research used resources of the Oak Ridge Leadership Computing Facility, which is a US DOE Office of Science User Facility supported under Contract no. DE-AC05-00OR22725.

The authors would like to thank Roger J. Kapsimalis for insightful discussions on the work, the developers of RMG for user assistance and access to their RMG-Phonopy interface, and Y. Q. Cheng for assistance with OCLIMAX.

Data availability

The data presented in this manuscript is available upon request from the corresponding author.

References (43)

  • P.W. Albers et al.

    Neutron scattering study of the terminating protons in the basic structural units of non-graphitising and graphitising carbons

    Carbon

    (2016)
  • M. Casartelli et al.

    Structure and stability of hydrogenated carbon atom vacancies in graphene

    Carbon

    (2014)
  • K.F. Garrity et al.

    Pseudopotentials for high-throughput DFT calculations

    Comput. Mater. Sci.

    (2014)
  • S.B. Isbill et al.

    Reviewing computational studies of defect formation and behaviors in carbon fiber structural units

    Comput. Mater. Sci.

    (2021)
  • V.N. Popov et al.

    Resonant Raman spectra of graphene with point defects

    Carbon

    (2009)
  • H. Kim et al.

    The stretching vibration of hydrogen adsorbed on epitaxial graphene studied by sum-frequency generation spectroscopy

    Chem. Phys. Lett.

    (2011)
  • E. Frank et al.

    Carbon fibers: precursor systems, processing, structure, and properties

    Angew. Chem. Int. Ed. Engl.

    (2014)
  • R. Böhm et al.

    Reinforcement systems for carbon concrete composites based on low-cost carbon fibers

    Fibers

    (2018)
  • D.L. Chung

    Carbon Fiber Composities

    (1994)
  • X. Huang

    Fabrication and properties of carbon fibers

    Materials

    (2009)
  • T. Matsumoto

    Mesophase pitch and its carbon fibers

    Pure Appl. Chem.

    (1985)
  • Notice: This manuscript has been authored by UT-Battelle, LLC, under contract DE-AC05-00OR22725 with the US Department of Energy (DOE). The US government retains and the publisher, by accepting the article for publication, acknowledges that the US government retains a nonexclusive, paid-up, irrevocable, worldwide license to publish or reproduce the published form of this manuscript, or allow others to do so, for US government purposes. DOE will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan (https://energy.gov/downloads/doe-public-access-plan).

    View full text