Graphite Melting Research Articles

Monotropic type transformations of diamond below the Berman-Simon line

We construct the phase diagram of a monotropic type for carbon under the Berman-Simon line. It describes metastable phase transformations occurring in diamond when the graphitization is delayed. However, thus far, monotropic phase transition of metastable diamond could not be experimentally observed. To produced delay conditions, we introduced at diamond surface a droplet of graphite melt with a temperature of about 4800K. We demonstrate that thermal shock provided mechanical stress and competition between phase transformations occurring at two high temperature and high pressure branches. At the so called “fast graphitization line” (reported by F.P. Bundy et al., Carbon 34 (1996) 141) the transformations take place tending to the spontaneous diamond graphitization, whereas at the line of monotropic phase diagram, they tend toward the melting of diamond. The transition can start as the graphitization, whereas change-over to the second branch can occur as a result of rapid unloading of mechanical stresses. The melting wins this is competition due to its faster kinetics. We have also found out the characteristic features underlying the initiation of processes near the “fast graphitization line” as are dealing with the spontaneous formation in diamond of a network of compressed graphite, and with the melting of such graphite in the course of its expansion taking place owing to the loss of mechanical stability of diamond caused by the network. This mechanism, apparently reveals a liquid-like nature of the transformations at the “fast graphitization line”.

Atomistic structure and anomalous heat capacity of low-density liquid carbon: Molecular dynamics study with machine learning potential

Liquid carbon remains the source of several unsolved questions related to its structure and region of thermodynamic stability. Experiments demonstrate a drastic decrease in the density for liquid carbon along the graphite melting line in the pressure range P = 1–3 GPa and the nature of this phenomenon is unclear. A recent experimental study [A.M. Kondratyev and A.D. Rakhel, PRL (2019)] revealed another peculiar and still unexplained feature of the liquid carbon – its excessive heat capacity. Using classical molecular dynamics with machine learning potential GAP-20, we study the structural properties of liquid carbon and demonstrate that at P < 1–2 GPa it resembles a net of sp-hybridized chains, rather than a typical covalent liquid, with nanoscale porosity of this phase being responsible for the density decrease. We also show that excessive heat capacity could be a direct manifestation of a smooth transition from a high-density sp2-hybridized phase into a low-density sp-hybridized.

Graphite melting line

The study was performed to establish a graphite melting line. Bundy’s phase diagrams are presented, in which the melting point of graphite (Tm) does not exceed 5500 K for all the considered pressures, up to (at least) 1 Mbar. The results of several studies on pulsed current heating related to measuring the melting temperature of graphite and obtaining the liquid phase of carbon are discussed. It is indicated that the Tm « 6300 K was obtained at a pulse pressure of 38-50 kbar (estimates) in 1999 year. The results of pulsed heating of graphite with current pulse for 1µs duration and obtaining Tm« 6400 K with calculation of pressure « 3-18 kbar in 2019 are considered. Possible errors of this last experiment are noted.

Comparative study of melting of graphite and graphene

Abstract The melting lines of graphite and liquid carbon have been studied for a long time. However, numerous controversies still remain in this field; for instance, different experiments give different melting temperatures. In this work, we explore the melting lines of graphite and graphene by means of classical and ab-initio molecular dynamics. We show that the empirical models developed on the basis of experimental data (AIREBO potential, Tersoff potential and some others) fail to reproduce the properties of liquid carbon properly. The models fitted to ab-initio data (LCBOPII and GAP) give much better results. However, both types of empirical models and ab-initio simulations evidence the presence of smooth structural crossover in liquid carbon. We also show that the “melting” of graphene discussed in previous works on computer simulation is indeed sublimation and propose a novel method to simulate the melting of graphene: simulation of graphene sheet in argon atmosphere. The melting temperature of graphene in argon atmosphere appears to be close to the melting temperature of graphite.

Investigation of the microstructure change due to phase transition in nanosecond pulsed laser processing of diamond

Experiments and theory are employed to investigate the thermal damage induced by infra-red nanosecond pulses in atmospheric air into a boron-doped diamond target. Micro-Raman spectroscopy, Transmission Electron Microscopy (TEM) analysis and surface topography measurement are used to investigate the carbon phase created during the rapid heating and cooling of diamond, as well as the amount of material ablated during the interaction with the laser. The analysis provides insight into the phenomena occurring for the rapid graphitisation of diamond during pulsed laser ablation, and also the microstructural disorder induced by the thermal and pressure fields at level of energy below and above the melting threshold. To support the understanding from the experimental investigations, a model is constructed for the graphitisation and ablation of diamond coupled with a collisional radiative model for the plasma evolution. The one-dimensional system of non-linear equations that model the physical processes provides an insight into the dynamics of the phenomena leading to the creation of disturbed graphite during pulsed laser ablation. Furthermore, the model helps to identify the main physical processes leading to the creation of disordered graphite, suggesting that plasma evolution does not follow a Boltzmann-Saha equilibrium and that radiative recombination is a main factor influencing the thermal evolution of the plasma and the diamond target. Finally, a good agreement with experimental findings is obtained, particularly in regards to the amount of material ablated, the thickness of the graphite layer and the processes leading to the melting of graphite.

Thermophysical properties of graphite HOPG and HAPG in the solid state and under melting (from 2000 K up to 5000 K)

Experiments with HOPG graphite grade showed that the melting temperature of graphite equals 4800-4900 K and that the melting of graphite is possible only at elevated pressures. The data were obtained for resistivity, specific heat and input (Joule) energy up to 5000 K. HAPG (Highly Annealing Pyrolytic Graphite) is a form of highly oriented pyrolytic graphite. HAPG specimens in the form of strips (thickness 30 microns) were placed in a cell (between two plates of glass-sapphire). The specimen temperature was measured by a high speed pyrometer. The heat of fusion for both graphite grades (heated in a confined volume) was less (and specific heat – higher) than for the case with nearly free expansion. A possible reason for the observed effects is discussed in the report.

Peculiarities of high-temperature properties of HAPG graphite within the melting region

By means of pulse current heating (3–5 μs), specimens of a new highly oriented pyrolytic graphite (the highly annealing pyrolytic graphite, HAPG) within the temperature range 2500–5000 K were investigated. The temperature, electrical resistance (compared to the initial sizes), and Joule energy of heating were measured. The beginning of graphite melting was registered at 4800–5000 K. It was observed that HAPG is characterized by a lower melting heat as compared to other types of graphite.

Experimental study of liquid carbon

Direct measurements of the functional dependencies of the electric resistivity and the molar volume on enthalpy and pressure have been performed for graphite and liquid carbon. It has been found that for graphite at the pressures P ⩽ 1 GPa the isochoric temperature coefficient of resistance is positive, while for liquid carbon it is negative over the entire pressure range investigated where P = 0.5–3.5 GPa. These observations probably indicate that graphite is a metal whereas liquid carbon is not a metal, so that the melting of graphite under such pressures coincides with a metal-to-nonmetal transition.

Molecular-dynamics based insights into the problem of graphite melting

The experimental data on graphite melting temperature remain poorly determined despite the long history of investigations. The experimental results cover the wide span from 3800 to 5000 K that is an essentially larger uncertainty than the errors of individual experiments. In this work, we deploy the molecular dynamics method and expand our previous study of the kinetics of graphite melting comparing different carbon interatomic potentials. Here we consider the melting front propagation rate, the aspects of defect formation and single graphene layer decay. The results obtained allow us also to discuss the aspects of graphite-vapor and possible graphite-carbyne phase transitions at low pressures.

Quenching of liquid carbon under intensive heat transfer to the cold diamond substrate: Molecular-dynamic simulation

Quenching of liquid carbon (T = 6600 K) on a cold diamond substrate at T = 300 K in conditions close to the experimental laser melting of dispersed graphite on the substrate of natural diamond is investigated using molecular dynamics (MD) simulations. Quenching was carried out for two types of boundary conditions on the side opposite to the diamond substrate. The simulations confirmed the experimental result of the formation of amorphous carbon under such conditions. The calculations showed that the destruction of the diamond substrate did not take place because of its very high thermal conductivity. The estimation of the cooling rate of liquid carbon was done, the result is 1015 K/s. Temperature profiles in different layers of liquid carbon were restored to reproduce the detailed picture of the quenching process. We evaluated the radial distribution functions (RDF), the distribution of carbon atom bond fractions sp1-sp2-sp3, the average bond length and the azimuthal angles distributions for amorphous carbon atoms. This analysis confirmed that the amorphous carbon obtained by quenching in MD-simulations had a graphite-like structure.

Graphite melting: Atomistic kinetics bridges theory and experiment

The graphite melting temperature remains poorly determined despite the considerable effort accomplished since the work of Bundy (1963) [1]. The absence of a consensus on its melting temperature at normal conditions has been considered as a technical problem that motivated more and more sophisticated experiments. The experimental evidences of the maximum on the graphite melting curve resulted in the liquid–liquid phase transition hypothesis for liquid carbon. However this hypothesis still requires a sound evidence. In this work using atomistic methods we focus on the kinetics of graphite melting and show that the experimental puzzles can be resolved by considering the graphite melting as a process in the non-equilibrium superheated solid. The unusually slow melting kinetics results in the existence of the superheated graphite at the microsecond timescale and thus biases the measurements of its equilibrium melting temperature.

Molecular dynamics simulation of graphite melting

Questions on the behavior of the graphite melting curve have remained open during the last fifty years. The process of graphite melting in the pressure range of 2–14 GPa is investigated by the method of molecular dynamics using the model of reactive interatomic potential; the dynamics of melting-front propagation upon crystal superheating is considered, and the melting curve is plotted. The self-diffusion coefficient in the liquid phase is determined for the aforementioned pressure range, and the question of the existence of the liquid-liquid phase transition in carbon is considered.

Isotropic graphite melting under pulsed heating

Experiments on pulsed heating (few microseconds) of graphite with measurements of the liquid carbon resistivity are described. It is confirmed that heating in water at atmospheric pressure do not allow production and study of liquid carbon; in the best case, the liquid state region beginning is achieved. Heating in sapphire tubes results in pulsed pressure (to ten of kbar) as expanding graphite bears against the tube wall. This increasing (during few microseconds) pressure makes it possible to study the carbon liquid state in a limited volume. Isochoric heating resulted in the possibilities ofmeasuring the liquid carbon resistivity at high specific energies (to ∼32 kJ/g) and high pressures. Such measurements are extremely expensive at stationary studies.

Properties of graphite at melting from multilayer thermodynamic integration

Although the melting of graphite has been experimentally investigated for a long time, there is still much debate on the graphite melting properties, as studies show significant discrepancies. We calculate the melting line by means of LCBOPII, a state-of-the-art interaction potential for carbon. To this purpose, we developed a generalized thermodynamic integration scheme, suitable for layered crystals. We also investigate the structure of liquid carbon around the coexistence line, including the undercooled region, as well as its dynamic properties in a wide range of temperatures.

Reproducing the Shilla Dynasty’s Direct-Bonding Granulation Process

Granulation is a precious metal craft process method that decorates a metal surface using tiny metal granules. It was imported into Korea during the Shilla Dynasty around 1500 years ago, and many granulation ornaments have been found with the process’s unique bonding features. The granules show a direct bonded interface with a neck. The key technology of making granules and bonding the granules is not well known. Thus, it is a technology of the Lost World. Although the exact bonding method is unidentifiable, it is known that the traditional method of preparing gold granules was time consuming and costly. Therefore, we proposed a process to reproduce the Shilla’s granulation ornament using a modern method. First, we employed atomization to produce 22K gold granules. Direct bonding was accomplished using a spot welder and vacuum jig instead of using the traditional method of graphite bed melting and direct annealing. 0.8 mm granules were successfully fabricated and bonded directly to the substrate with a necking and 35% bonding ratio, which is very similar to Shilla’s granule bonding. Moreover, to estimate the bond strength, K factors (fracture toughness index) at different bonding ratios were evaluated using a finite element method simulation. Our proposed direct bonded granule process and design were expected to have enough bond strength to be used as a key element for fine modern jewelry.

Regeneration of graphite explosive-emission cathodes operating at high repetition rates of nanosecond accelerating pulses

We have experimentally studied the possibility of increasing the working life of explosive-emission cathodes operating at high repetition frequencies (several kilohertz) of nanosecond accelerating voltage pulses. Micrographs of the cathode surface upon training, photographs of the operating cathode, and analyses of the material carried away from the cathode showed that a substantial role in the cathode degradation is played by heating of some regions of the emissive edge up to a temperature of graphite melting.

Modeling the Phase Diagram of Carbon

We determined the phase diagram involving diamond, graphite, and liquid carbon using a recently developed semiempirical potential. Using accurate free-energy calculations, we computed the solid-solid and solid-liquid phase boundaries for pressures and temperatures up to 400 GPa and 12 000 K, respectively. The graphite-diamond transition line that we computed is in good agreement with experimental data, confirming the accuracy of the employed empirical potential. On the basis of the computed slope of the graphite melting line, we rule out the hotly debated liquid-liquid phase transition of carbon. Our simulations allow us to give accurate estimates of the location of the diamond melting curve and of the graphite-diamond-liquid triple point.

Understanding and predicting the temporal response of laser-induced incandescence from carbonaceous particles

This paper describes a model for analyzing and predicting the temporal behavior of laser-induced incandescence (LII) from combustion-generated soot, carbon black, and other carbonaceous particles on a nanosecond time scale. The model accounts for particle heating by absorption of light from a pulsed laser and cooling by sublimation, conduction, and radiation. The model also includes mechanisms for oxidation, melting, and annealing of the particles and nonthermal photodesorption of carbon clusters from the particle surface. At fluences above 0.1 J/cm2, particle temperatures during the laser pulse are determined by the balance between absorption and sublimation, whereas at lower fluences particle temperatures do not reach the sublimation temperature, and temperatures are predominantly controlled by absorption and conduction. After the laser pulse, temperatures are predominantly controlled by conductive cooling rates. Oxidative heating may compete with conductive cooling on these time scales. Annealing of the particles to a more ordered phase of carbon is predicted to occur at fluences as low as 0.02 J/cm2. Annealing may strongly influence sublimation rates, and changes in emissivity during annealing are predicted to increase signal decay rates. Supersonic expansion of the carbon clusters sublimed from the surface is calculated to occur at fluences above 0.12 J/cm2. When compared with LII measurements recorded in a flame at atmospheric pressure, the model reproduces the shapes and relative magnitudes of LII temporal profiles over a wide range of laser fluences. Comparisons between model predictions and experimental observations suggest that the particles do not melt at laser fluences that lead to melting of bulk graphite. These comparisons also indicate that the energy released during particle annealing is much smaller than that released during annealing of neutron- or electron-irradiated graphite. Despite good agreement between model and experimental results, large uncertainties exist for input parameters used to calculate annealing rates and rates of oxidation, conduction, absorption, emission, and photolytic desorption of carbon clusters for both the initial and annealed particles.

Theoretical approach to the laser-induced melting of graphite under different pressure conditions

We present a theoretical study of the laser-induced femtosecond melting of (1) graphite under high external pressure and (2) ultrathin graphite films under normal conditions. Our approach consists of molecular dynamic simulations performed on the basis of a time-dependent, many-body potential energy surface derived from a tight-binding Hamiltonian. Our results show that the laser-induced melting process occurs in two steps: (i) destruction of the graphite sheets via bond breaking, and (ii) merging of the melted layers. The separation of the two steps is more evident for graphite under pressure (10 GPa), but is also present in graphite films at normal pressure. The melting product is a low-density carbon phase, which remains stable under high pressure, but is unstable with an ultrashort life-time under normal pressure.

Liquid-liquid phase transition in elemental carbon: a first-principles investigation.

It has been recently suggested that elemental carbon may be a promising candidate to exhibit a liquid-liquid phase transition (LLPT). We report the results of first-principles molecular dynamics simulations showing no evidence of LLPT in carbon, in the same temperature and pressure range where such a transition was found using empirical calculations. Our simulations indicate a continuous evolution from a primarily sp-bonded liquid to an sp(2)-like and an sp(3)-like fluid, as a function of pressure, above the graphite melting line. The discrepancy between quantum and classical simulations is attributed to the inability of empirical potentials to describe complex electronic effects in condensed carbon phases.