ArticlePDF Available

# Shock Compression of Liquid Carbon Monoxide and Methane to 90 GPa (900 kbar)

Authors:

## Abstract and Figures

Dynamic equation-of-state data for liquid CO and CHâ were measured in the shock pressure range 5--92 GPa (50--920 kbar) using a two-stage light-gas gun. The liquids were shocked from initial states near their saturation curves at 77 and 111 K for CO and CHâ, respectively. The experimental technique used to double-shock CHâ is described. The CO data were examined by using three theoretical models: (1) a chemically nonreactive model, (2) a quasi-chemical-equilibrium model that allows CO to dissociate into gaseous species and graphite, and (3) a chemical-equilibrium model that also includes a dense carbon phase which exists at higher pressures and temperatures than graphite. This dense phase is assumed to be diamond. Our analysis shows that at low pressure chemical equilibrium takes much longer than a typical shock passage time. As a consequence, the experimental data initially follow the nonreactive Hugoniot to pressures well beyond the chemical dissociation limit. Both the experimental data and the Hugoniot computed with case (3) agree satisfactorily at high pressure. Further consequences of these observations to high-explosive studies are discussed. The theoretical analysis for the CHâ data was presented in an earlier paper.
Content may be subject to copyright.
A preview of the PDF is not available
... More recently, Leonhardi performed AIMD simulations to characterise CO fluid to 140 GPa and 5000 K and obtained an EOS (Leonhardi-2017 EOS) (Leonhardi and Militzer, 2017). Several shockwave experiments exist on CO and CO 2 , potentially allowing an assessment of the various EOSs, however little accurate temperature determination exists (Nellis et al., 1991;Nellis et al., 1981). Notably, in the shock experiments, CO and CO 2 molecules decomposed at high P and condensed into the diamond phase. ...
... For CO fluids, we found one set of EOS data from classical MD simulations (BS-1991) (Belonoshko and Saxena, 1991b), and one set of EOS data based on AIMD simulations (Leonhardi-2017) (Leonhardi and Militzer, 2017). We plot them in Fig. 6 together with our results and the shock-wave data from Nellis et al. (1981). It was shown by Nellis's experiment (Nellis et al., 1981) that the CO fluids with molar volume less than 13 cm 3 /mol are all deeply polymerised or solidified. ...
... We plot them in Fig. 6 together with our results and the shock-wave data from Nellis et al. (1981). It was shown by Nellis's experiment (Nellis et al., 1981) that the CO fluids with molar volume less than 13 cm 3 /mol are all deeply polymerised or solidified. Nellis also suggested a dense solid carbon phase at these conditions. ...
Article
Ab initio molecular dynamics simulations were performed at pressures and temperatures up to 160 GPa and 4000 K, in order to obtain equations of state (EOS) for CO and CO2 fluids. We found that polymerisation of CO and CO2 fluids starts at low pressures, and that including the effect of polymerisation is essential for accurate EOS. EOSs for CO and CO2 determined from methods using experimental data, or classical potentials that ignore the changes in speciation, should be treated with caution when extrapolated beyond the examined pressures and temperatures. The obtained data was fitted to a modified Lee and Kesler EOS for both CO and CO2 fluids. The thermodynamic calculations for the decarbonation reactions of both MgCO3 and CaCO3 using the derived CO2 EOS reproduced the experimental data and theoretical calculations at low pressures. Both MgCO3 and CaCO3 pure phases are found to be stable in the upper mantle compared to CO2. However, they both become destabilised when approaching lower mantle conditions.
... C/O systems are often employed in practical applications and are excellent candidates for uncovering the essential features of this complex process due to their relative simplicity. For example, carbon monoxide (CO) has been used for carbon nanotube production under ambient to slightly elevated pressures, and explored as a precursor for other nanocarbon materials [16][17][18] , while shock compression of liquid CO to 10s of GPa has yielded convincing evidence of nanocarbon formation 6,19 . Furthermore, C/O systems serve as a logical stepping stone to the CNO (and CHNO) organic compounds traditionally employed for high-pressure nanodiamond synthesis (e.g., benzotrifuroxan, C 6 N 6 O 6 ) 8 , or those recently shown to yield carbon nano-onions (e.g., bis(nitrofurazano)furazan, C 6 N 8 O 8 ) 7 . ...
Article
Full-text available
There is significant interest in establishing a capability for tailored synthesis of next-generation carbon-based nanomaterials due to their broad range of applications and high degree of tunability. High pressure (e.g., shockwave-driven) synthesis holds promise as an effective discovery method, but experimental challenges preclude elucidating the processes governing nanocarbon production from carbon-rich precursors that could otherwise guide efforts through the prohibitively expansive design space. Here we report findings from large scale atomistically-resolved simulations of carbon condensation from C/O mixtures subjected to extreme pressures and temperatures, made possible by machine-learned reactive interatomic potentials. We find that liquid nanocarbon formation follows classical growth kinetics driven by Ostwald ripening (i.e., growth of large clusters at the expense of shrinking small ones) and obeys dynamical scaling in a process mediated by carbon chemistry in the surrounding reactive fluid. The results provide direct insight into carbon condensation in a representative system and pave the way for its exploration in higher complexity organic materials. They also suggest that simulations using machine-learned interatomic potentials could eventually be employed as in-silico design tools for new nanomaterials.
... For example, carbon monoxide (CO) has been used for carbon nanotube production under ambient to slightly elevated pressures, and explored as a precursor for other nanocarbon materials, [16][17][18] while shock compression of liquid CO to 10s of GPa has yielded convincing evidence of nanocarbon formation. 6,19 Furthermore, C/O systems serve as a logical stepping stone to the CNO (and CHNO) organic compounds traditionally employed for high-pressure nanodiamond synthesis (e.g. benzotrifuroxan, C 6 N 6 O 6 ), 8 or those recently shown to yield carbon nano-onions (e.g. ...
Preprint
Full-text available
There is significant interest in establishing a capability for tailored synthesis of next-generation carbon-based nanomaterials due to their broad range of applications and high degree of tunability. High pressure (e.g. shockwave-driven) synthesis holds promise as an effective discovery method, but experimental challenges preclude elucidating the processes governing nanocarbon production from carbon-rich precursors that could otherwise guide efforts through the prohibitively expansive design space. Here we report findings from large scale atomistically-resolved simulations of carbon condensation from C/O mixtures subjected to extreme pressures and temperatures, made possible by machine-learned reactive interatomic potentials. We find that liquid nanocarbon formation follows classical growth kinetics driven by Ostwald ripening (i.e. growth of large clusters at the expense of shrinking small ones) and obeys dynamical scaling in a process mediated by carbon chemistry in the surrounding reactive fluid. The results provide direct insight into carbon condensation in a representative system and pave the way for its exploration in higher complexity organic materials. They also suggest that simulations using machine-learned interatomic potentials could eventually be employed as in-silico design tools for new nanomaterials.
... For water, we use the polynomial approximation of the EOS determined by Ree (1976), as well as the ab initio EOS of Mazevet et al. (2019). Due to their smaller expected abundances, the accuracy for CH 4 and NH 3 is less crucial, and again, we use the polynomial approximations given by Hubbard et al. (1995) for the shockwave CH 4 data determined by Nellis et al. (1981) as well as the approximation given by Hubbard et al. (1995) for the zero-temperature CH 4 equation of state. Finally, to model the density of rock within the planets, we employ the EOS from used by Hubbard & Macfarlane (1980) for the mixture of 38% SiO 2 , 25% MgO, 25% FeS, and 12% FeO which we similarly take to approximately constitute "rock." ...
Article
Full-text available
Interior models of Uranus and Neptune often assume discrete layers, but sharp interfaces are expected only if major constituents are immiscible. Diffuse interfaces could arise if accretion favored a central concentration of the least volatile constituents (also incidentally the most dense); compositional gradients arising in such a structure would likely inhibit convection. Currently, two lines of evidence suggest possible hydrogen-water immiscibility in ice giant interiors. The first arises from crude extrapolation of the experimental hydrogen-water critical curve to ∼3 GPa (Bali et al. 2013). The data are obtained for an impure system containing silicates, though Uranus and Neptune could also be "dirty." Current ab initio models disagree (Soubiran & Militzer 2015), though hydrogen and water are difficult to model from first-principles quantum mechanics with the necessary precision. The second argument for hydrogen-water immiscibility in ice giants, outlined herein, invokes reasoning about the gravitational and magnetic fields. While consensus remains lacking, here we examine the immiscible case. Applying the resulting thermodynamic constraints, we find that Neptune models with envelopes containing a substantial water mole fraction, as much as χ≳0.1 relative to hydrogen, can satisfy observations. In contrast, Uranus models appear to require χ≲0.01, potentially suggestive of fully demixed hydrogen and water. Enough gravitational potential energy would be available from gradual hydrogen-water demixing, to supply Neptune's present-day heatflow for roughly ten solar system lifetimes. Hydrogen-water demixing could slow Neptune's cooling rate by an order of magnitude; different hydrogen-water demixing states could account for the different heatflows of Uranus and Neptune.
... For water, we use the polynomial approximation of the EOS determined by Ree (1976), as well as the ab initio EOS of Mazevet et al. (2019). Due to their smaller expected abundances, the accuracy for CH 4 and NH 3 is less crucial, and again, we use the polynomial approximations given by Hubbard et al. (1995) for the shockwave CH 4 data determined by Nellis et al. (1981) as well as the approximation given by Hubbard et al. (1995) for the zero-temperature CH 4 equation of state. Finally, to model the density of rock within the planets, we employ the EOS from used by Hubbard & Macfarlane (1980) for the mixture of 38% SiO 2 , 25% MgO, 25% FeS, and 12% FeO which we similarly take to approximately constitute "rock." ...
Preprint
Interior models of Uranus and Neptune often assume discrete layers, but sharp interfaces are expected only if major constituents are immiscible. Diffuse interfaces could arise if accretion favored a central concentration of the least volatile constituents (also incidentally the most dense); compositional gradients arising in such a structure would likely inhibit convection. Currently, two lines of evidence suggest possible hydrogen-water immiscibility in ice giant interiors. The first arises from crude extrapolation of the experimental hydrogen-water critical curve to $\sim 3$ GPa (Bali et al. 2013). The data are obtained for an impure system containing silicates, though Uranus and Neptune could also be "dirty." Current ab initio models disagree (Soubiran & Militzer 2015), though hydrogen and water are difficult to model from first-principles quantum mechanics with the necessary precision. The second argument for hydrogen-water immiscibility in ice giants, outlined herein, invokes reasoning about the gravitational and magnetic fields. While consensus remains lacking, here we examine the immiscible case. Applying the resulting thermodynamic constraints, we find that Neptune models with envelopes containing a substantial water mole fraction, as much as $\chi \gtrsim 0.1$ relative to hydrogen, can satisfy observations. In contrast, Uranus models appear to require $\chi \lesssim 0.01$, potentially suggestive of fully demixed hydrogen and water. Enough gravitational potential energy would be available from gradual hydrogen-water demixing, to supply Neptune's present-day heatflow for roughly ten solar system lifetimes. Hydrogen-water demixing could slow Neptune's cooling rate by an order of magnitude; different hydrogen-water demixing states could account for the different heatflows of Uranus and Neptune.
... Nellis et al. (11)(12)(13)(14) measured the rates of dissociation of N2, CO2, and CO plus ionization of H2O under high shock pressures. However, the shock temperatures reached for the individual diatomic or triatomic molecules are considerably higher than those of reaction product mixtures at the same shock pressures. ...
Conference Paper
The Jones-Wilkins-Lee (JWL) equation of state (EOS) has long been used to accurately calculate the Chapman - Jouguet (C-J) state of condensed phase explosive detonation waves and the subsequent expansion of the reaction products as they do work on surrounding materials. In many applications, the states of the reaction products must also be known at higher pressures and temperatures than the C-J state. Such states occur in overdriven detonation waves supported by high velocity impacts, reflected waves, and converging waves. When the states attained in overdriven detonations were first measured experimentally, the initial JWL EOS's based only on expansion data were found to be too compressive. This problem was resolved 30 years ago by modifying the two exponential terms in the JWL EOS to yield less compressible states at higher pressures, while still matching C-J state and product expansion states at lower pressures. The experimental data on overdriven detonation waves in Pentaerythritol tetranitrate (PETN) is used to develop accurate JWL EOS's using three methods. The first method is to use the analytical formulas of Urtiew and Hayes, who used the experimental C-J detonation velocity, C-J pressure, and the heat of reaction to develop JWL EOS's that fit high and low pressure data. The second method is to use the CHEETAH chemical equilibrium code to calculate: the C-J state; the adiabatic expansion of the products; and a JWL EOS fit to its predicted expansion states. The accuracy of the CHEETAH calculated PETN C-J detonation velocities was checked against experimental detonation velocities measured at several initial densities ranging from 0.27 g/cm³ to 1.764 g/cm³. The third method is to use the CHEETAH code to calculate the overdriven Hugoniot states based on its C-J calculation. This method accounts for the changes in reaction product concentrations as the shock pressures and temperatures increase. Excellent agreement with the experimental shock pressures and densities approaching 120 GPa and 3.6 g/cm³ was obtained using all three methods.
Article
We present measurements of diamond formation in doubly shocked Stycast 1266 epoxy (comprising C, H, Cl, N, and O) using in situ x-ray diffraction. Epoxy samples were reshocked against a LiF window to pressures between 80 and 148 GPa in experiments at the Omega Laser Facility. The pressure and temperature conditions were diagnosed in situ using velocimetry and optical pyrometry, respectively. X-ray diffraction patterns of the compressed epoxy are consistent with cubic diamond (Fd[Formula: see text]m), indicating that diamond can precipitate not only from twice-shocked CH polystyrene [Kraus et al. Nat. Astron. 1, 606 (2017)] at these conditions but also from twice-shocked CH polymers with the addition of oxygen, nitrogen, and chlorine. These results, in combination with previous works on CH, [Formula: see text], [Formula: see text], and methane hydrate, support that diamond often, but not always, forms from CH-based compounds at extreme pressures and temperatures, indicating that the chemical composition, thermodynamic compression path, and kinetics play an important role.
Article
Full-text available
Molecular dynamics simulation of the pressure-density-temperature properties of supercritical methane (CH4) are made with the COMPASS II force field model in the range of 200–3000 K, 0.1–3.0 GPa, and 0.22–0.668 g·cm⁻³, where 710 states are simulated using NPT ensemble, and 212 states are simulated using NVT ensemble. These results are in good agreement with experimental data and the calculated results from highly accurate reference model of Setzmann and Wagner (J Phys Chem Ref Data 20:1061–1155, 1991) and its extrapolation in the region where the reference model can be validated. The simulation results are calibrated with the reference model. The calibrated simulations results and the reference model are used simultaneously to develop an accurate cubic equation of state for supercritical CH4 in the range of about 300–3000 K and 0–3 GPa (0–0.53 g·cm⁻³). The equation are tested against experimental and simulated data at high temperatures and pressures. Compared with the overwhelming majority of experimental results, the volume deviations are within 0.4 % to 1.1 %, with averages of about 0.1 % to 0.4 %; Compared with the molecular simulation results in literature and this work, the volume deviations are within 0.6 % to 3.7 %, with averages of about 0.1 % to 1.2 %. The equation can accurately predict the fugacity coefficients, residual enthalpies, and entropies and other thermodynamic properties.
Article
We describe a machine learning approach to rapidly tune density functional tight binding models for the description of detonation chemistry in organic molecular materials. Resulting models enable simulations on the several 10s of ps scales characteristic to these processes, with “quantum-accuracy.” We use this approach to investigate early shock chemistry in 3,4-bis(3-nitrofurazan-4-yl)furoxan, a hydrogen-free energetic material known to form onion-like nanocarbon particulates following detonation. We find that the ensuing chemistry is significantly characterized by the formation of large CxNyOz species, which are likely precursors to the experimentally observed carbon condensates. Beyond utility as a means of investigating detonation chemistry, the present approach can be used to generate quantum-based reference data for the development of full machine-learned interatomic potentials capable of simulation on even greater time and length scales, i.e., for applications where characteristic time scales exceed the reach of methods including Kohn–Sham density functional theory, which are commonly used for reference data generation.
Article
Full-text available
Shockwave compression of methane was simulated using the molecular dynamics method with Hugoniostat and MSST approach. The calculation was carried out employing ReaxFFlg potential. We show comparison of simulated Hugoniots for pressure range 0,1-44 GPa and times required for simulation run of two different methods. We also compare two ReaxFF potentials: ReaxFF(2008) and ReaxFF-lg
Article
Full-text available
A variational technique which is based on two different inequalities for the Helmholtz free energies is used to calculate the equilibrium thermodynamic properties of simple fluids. A system with hard‐sphere potential function is used as the reference system. Helmholtz free energy of the original system is calculated by variation around the Helmholtz free energy of the reference system, and the other thermodynamic properties are calculated from free energy. By choosing a hard‐sphere reference system, it is possible to calculate the equilibrium thermodynamic properties of fluids from very low densities to densities close to solid, and from high temperatures in the gas phase to low temperatures in the liquid phase, in the ranges where experimental and machine‐calculated data are available. It is shown that the present variational technique is a better approach to the prediction of the equilibrium thermodynamic properties of liquids and vapor–liquid phase transition than any other approach so far developed. While the variational calculation based on a hard‐sphere reference system does not predict the liquid–solid phase transition, it is argued that this might be due to the neglect of the orientation, or ordering in the formulation of the working inequality for fluids.
Article
Full-text available
Dynamic equation‐of‐state data for liquid Ar, N2, and O2 were measured in the shock pressure range 21–91 GPa (210–910 kbar) by means of a two‐stage light‐gas gun. The liquids were shocked from initial states near their saturation curves at 0.1 MPa (1 bar) and 80 K. The cryogenic target system is described. The initial liquid densities were obtained by measuring the temperature and pressure of the specimens and calculating the densities from published equations of state. Shock velocities were measured to 0.5–1.1% accuracy with an electronic detection system with subnanosecond time resolution. Impactor velocities in the range 4–7 km/s were measured to 0.1% accuracy with a flash radiographic technique. Mass velocities were obtained by the method of shock impedance matching. The data are discussed in terms of the statistical–mechanical theories of Ross and Ree.
Article
Shock wave compression curves for liquid argon were computed using a number of recently proposed pair potentials. The results are compared to experiment and used to evaluate the accuracy of the repulsive potential. Recent molecular beam results are found to be in excellent agreement with the shock data.
Article
A 33 term modified Benedict-Webb-Rubin equation of state is presented for methane. The adjustable parameters in the equation of state have been estimated using recent experimental data and least squares techniques which include the thermodynamic equilibrium conditions for the co-existing liquid and vapour phases. Comparisons of the new equation of state and an older modified Benedict-Webb-Rubin equation of state to experimental data are given.
Article
Electronegativity difference was redefined in Paper I of this series as a scaling parameter which combines the concepts of valence and size differences. A procedure has been developed for its evaluation in terms of a two-band model. In Paper II of this series it was shown that this model describes and predicts the ionization potentials and electronic interband gaps of binary ANB8-N compounds and their alloys. Here the energy of this model semiconducting-insulating solid is evaluated relative to a free-electron gas, i.e., an idealized metal, as a function of composition, pressure, and temperature. Using this highly simplified scaling approach, we obtain suprisingly accurate predictions for the heat of fusion, melting point, and pressure-temperature phase diagrams of these materials. A revised method of calculating the excess heat of mixing of a substitutional alloy is presented. This calculation is extended to the case of an arbitrary dilute impurity in an arbitrary semiconducting host; the distribution coefficient at the melting point of the host is obtained.
Article
Shock-induced graphite-to-diamond phase transition pressures and Hugoniot parameters to 120 GPa are presented for pyrolytic, Ceylon natural, and synthetic graphites. The response of amorphous carbon to shock loading, including a shock-induced phase transition, is reported. Interpretations of discontinuities found in the shock-velocity versus particle-velocity plots are discussed in terms of the carbon phase diagram. Evidence of a solid-to-solid phase transition from diamond to a metallic state at very high pressure is described.
Article
A variational theory is developed that is accurate at normal liquid densities and densities up to 4 times that of the argon triple point. This theory uses the inverse 12th‐power potential as a reference system. The properties of this reference system are expressed in terms of hard‐sphere packing fractions by using a modified form of hard‐space variational theory. As a result of this ’’bootstrapping,’’ a variational procedure may be followed that employs the inverse 12th‐power system as a reference but uses the hard‐sphere packing fraction as the scaling parameter with which to minimize the Helmholtz free energy.
Article
Shock wave data for liquid Ar, Xe, N2, O2, CO2, CH4, and CO have been used to test the possibility that the repulsive pair potential of these fluids scale in accordance with the ’’Law of Corresponding States.’’ This is found to be approximately valid to a compression of about 2.5 times liquid density. The law has been applied to compressed mixtures of N2 and O2 and a theoretical prediction of the detonation velocity and pressure of liquid NO is found to be in excellent agreement with experiment.
Article
Recent dynamic compression data for carbon are discussed which suggests that metallic carbon has unexpected thermodynamics properties, both in having a lower density and energy of formation and a higher melting temperature. On the basis of these properties the diamond phase of carbon is now predicted to transform to a solid phase of metallic carbon at all pressures above the graphite triple point.