The Influence of Processing Conditions on Gas Transport and Thermal Properties of Graphite Foil Compressed from Exfoliated Graphite

Abstract

Graphite foil (GF) compressed from exfoliated graphite (EG) is a sealing material, which is used in nuclear energy and the chemical industry. The preparation of graphite foil is a complex process, which includes the intercalation of graphite, water washing, thermal exfoliation and pressing of intermediate products. The preparation conditions significantly influence the structure of the material and its physicochemical properties. Thus, the aim of work was to reveal the correlation between GF processing conditions, its crystalline structure, porosity and gas permeability as well as thermal stability. Sealability of the material is connected with low value of gas permeability, while thermal stability allows use of the material in high-temperature processes. Optimization of these parameters allow for the obtaining of a reliable material and expanding of the areas of its application. Exfoliated graphite for GF was prepared at different temperatures of 600, 800 and 1000 °C from the H₂SO₄–graphite intercalation compound (GIC) of II, III, IV stages. The influence of the GF processing conditions (the GIC stage number and the EG preparation temperature) on the main properties (gas permeability and thermal oxidation stability) of the sealing materials was investigated. A decrease in GIC stage number leads to the formation of GF with lower macroporosity and lower nitrogen and hydrogen permeability. However, an increase in GF surface area leads to an increase in the rate of GF oxidation by air oxygen. An increase in the EG preparation temperature from 800 to 1000 ^oC results in the formation of EG with a developed micro- and mesoporosity and increasing GF gas permeability. A decrease in EG preparation temperature down to 600 °C promotes the formation of new transport macropores in GF. The change of the EG preparation temperature has little effect on GF oxidation stability.

1. Introduction

The advancement of energy and industrial technology is impossible without carbon materials [1,2]. Graphite seals prevent the flow of gas or liquid through the gap between two surfaces in pipelines, pumps, valves and other moving and fixed parts of industrial devices [3]. The seals provide the stable and uninterrupted operation of equipment. A sealant failure leads to increasing leakage of hazardous substances, environmental pollution, a threat to human health and equipment downtime and economic losses [4,5,6]. Proper selection and installation of seals is especially important in such branches of industry as nuclear energy, oil and natural gas processing, chemical and metallurgical industries. All of these branches require high performance sealing materials with gas and liquid impermeability, high thermal stability and mechanical properties.

Graphite foil (GF) is a unique sealing material with all of these properties [7,8,9,10]. High stability in a wide range of temperatures and chemical resistance allow for using GF in both low and high temperature processes such as gas condensation and separation, oil refining, using overheated steam, etc. High mechanical strength and elasticity, ability to withstand a high pressure of media, low material creep of GF result in effective sealability of different connections [3,11]. Graphite foil is obtained by pressing and calendering exfoliated graphite (EG), which makes it suitable for the manufacture of sealing materials. In turn, EG preparation occurs in several stages. At the first stage, the synthesis of graphite intercalation compounds (GIC) of various stages is carried out by intercalation of strong Brønsted acids (HNO₃, H₂SO₄, HClO₄) into the interlayer space of graphite in the presence of an oxidizer [12,13,14,15]. The stage number corresponds to the number of graphite layers between two layers of intercalate and decreases with an increase in the amount of intercalated substance. Next, GIC water washing is carried out with the formation of expandable graphite, after which it is thermally exfoliated at a high temperature [16,17,18].

The GIC stage number and the EG preparation temperature are crucial parameters, which define the crystalline and porous structure of GF compressed from EG [11,12,19]. The GIC stage number characterizes the oxidation degree of the graphite matrix and the amount of intercalate in the graphite matrix. The higher the oxidation degree reaches, the lower the GIC stage number that is formed [13,20,21]. An increase in the oxidation degree of the graphite matrix leads to the formation of a defective structure of EG and GF with development microporosity and low crystallite size [12]. The change in EG preparation temperature at the same time leads to the formation of low sized carbon fragments in the structure of GF and their oxidation and formation of new micro- and mesopores in GF [3]. Thus, the GF processing conditions significantly influence the properties of the material.

For industrial application, high performance graphite materials must have low gas permeability and high thermal stability [22,23,24,25]. These two properties significantly depend on the crystalline structure, surface area and porosity of the material [23,25,26]. Despite of having a relatively high porosity GF has transport pores with low sizes of 1.5–2.5 nm and 4.5–6 nm and surface macropores are mainly closed and have no influence on the gas transfer through GF [27,28].

The oxidation of graphite by air oxygen begins at 400–500 °C with the formation of CO₂ and CO [29,30]. The CO₂/CO molar ratio depends both on the temperature and on the oxidation mechanism. According to some studies, the oxidation of all substances with a structure similar to that of graphite begins with the most vulnerable areas, which are often called active oxidation centers [31,32,33]. In graphite, such centers are the edges of the graphene layers and point defects of the graphene planes. The oxidation process is also influenced by many factors such as the degree of crystallinity, the size and shape of crystallites, the shape of the graphite particle, the presence of catalytic and inhibitory impurities, surface area and the number of defects [29,34,35,36].

Expanding the scope of application of sealing materials based on GF requires control of their gas permeability and thermal stability to create a material with the desired functional properties. It is known that the GF preparation conditions have the greatest influence on the structure of EG and GF. Thus, varying the conditions can allow for obtaining the material with the minimum gas permeability (i.e., maximum sealability) and high thermal stability. The ability to control these characteristics is therefore important in both developing new materials and improving existing ones. On the other hand, understanding which structural parameters affect these properties makes the mechanisms of gas transport and oxidation more clear. The aim of this work is to investigate the influence of the processing conditions on crystalline and porous structure of graphite foil, its gas permeability and thermal properties.

2. Material and Methods

2.1. Sample Preparation

Natural graphite with an average flake size of 200–300 μm (99.9 wt%), concentrated sulfuric acid (96 wt%, 1.84 g cm⁻³) and potassium dichromate were used as starting materials.

To synthesize II, III, IV stages GIC-H₂SO₄, graphite was mixed with K₂Cr₂O₇ and 96 wt% H₂SO₄ in the mass ratio of m (graphite):m(K₂Cr₂O₇):m(H₂SO₄) = 1:0.09:7.4; 1:0.06:7.4; 1:0.04:7.4 respectively. The mixture was stirred for two hours and obtained graphite bisulfate was washed by hot water (60–70 °C) in the mass ratio of m(H₂SO₄):m(H₂O)~1:7 on the porous glass filter with the formation of expandable graphite. Expandable graphite was dried at a temperature of 60 °C for 6 hours and mass gain (Δm, %) was measured.

A tube furnace was used for the preparation of exfoliated graphite. A fixed mass of expandable graphite entered the preheated tube reactor with flowing air. An instantaneous heating of expandable graphite and an equal-time exposure of about 10 seconds at fixed temperature of 600, 800 and 1000 °C provide exfoliation of all samples at identical conditions. Obtained EG was collected in a special container at the outlet of the reactor. EG exfoliation volume (V_EG mL g⁻¹) was calculated as the ratio of EG volume to EG mass. The EG yield (Y_EG, %) was calculated as the ratio of EG mass to the mass of expandable graphite.

The samples of graphite foil with a density of 1.00 ± 0.05 g cm⁻³ were obtained prepared by calendering of exfoliated graphite. In the present work, the samples of exfoliated graphite and graphite foil are denoted as EG-n-T and GF-n-T, where n is the GIC stage number (n = II, III, IV) and T is the EG preparation temperature (T = 600, 800, 1000).

2.2. Investigation Techniques

2.2.1. Characterization

The phase composition of the samples was determined by X-ray diffraction (XRD) analysis on a Rigaku Ultima IV diffractometer using a copper X-ray emission tube (CuK_α radiation, λ = 0.154 nm). CuK_α radiation is a double of diffraction lines caused by K_α1 (λ_Kα1 = 0.15405 nm) and K_α2 (λ_Kα2 = 0.15444 nm) radiation. The diffraction lines caused by K_α1 and K_α2 radiation have been separated. For graphite materials, splitting due to K_α1 and K_α2 has been observed on the 006 line. The Kα1 component of (006) peak, which was resolved on the K_α1 and K_α2 doublet, was used for measurement of L_c. A pseudo-Voigt function was used for fitting the diffraction lines. The crystallite size along the c axis (L_c, nm) of the obtained samples was calculated using the Scherrer equation [37]:

L_c = 0.91 λ/(β cosΘ)

(1)

where λ is the wavelength of the copper K_α1 X-ray radiation (0.15405 nm); Θ is the angle (the Bragg angle, the angle of the diffracted wave, the angle between the primary X-ray beam (with λ wavelength) and the family of lattice planes, with interplanar spacing d), which equals to a half of the angle 2Θ (the goniometer angle, the diffraction angle, the angle between the direction of incident beams and resulting diffracted beam) of the peak (006) position; $\mathsf{\beta }=\sqrt{{\mathsf{\beta }}_m^2-{\mathsf{\beta }}_{Si}^2}$, β_m is the measured half-height width of K_α1 component of (006) peak and β_Si is the measured half-height width of K_α1 component of (422) peak of silicon external standard (cubic crystal system and Fd3m space group).

The morphology of expandable graphite and exfoliated graphite was investigated by scanning electron microscopy (SEM) using a scanning electron microscope TESCAN VEGA3 LMU.

The investigation of the macroporous structure of the GF samples was carried out on a mercury porosimeter AutoPore 9605 (Micromeritics). The macropore volume (V_macro, cm³ g⁻¹) was calculated from the amount of mercury intruded at a pressure of 31 MPa, which corresponds to cumulative volume of pores with a characteristic width of more than 40 nm.

The investigation of the total volume of pores less than 50 nm (V_meso, cm³ g⁻¹) and BET surface area (S_BET, m² g⁻¹) was carried out on an ASAP 2010N (Micromeritics) Porosimetry system. GF samples were outgassed for 2 h at 150 °C prior to measurements. The micro- and mesopore volume was calculated from the largest amount of N₂ adsorbed at P/P₀~0.95, which corresponds to cumulative volume of pores with a characteristic width of less than 50 nm.

2.2.2. Gas Permeance Measurements

The gas permeance in a parallel direction to the GF pressing axis was measured by a differential method using an experimental set-up for gas permeability measurement described in detail in work [28]. A disk of GF with a density of 1 g cm⁻³, a height of 0.6 mm and a diameter of 74 mm was placed and hermetically fixed in the diffusion cell. GF divides the cell volume into two cavities, which are the reservoir and receiver. The flow of nitrogen or hydrogen was supplied to the reservoir. The gas permeated though graphite foil was mixed with the carrier gas (helium) flow into the receiver. The concentration of the investigated gas in the mixture with the carrier gas was analyzed by a gas chromatograph. The flow rate of mixed gases was measured using a gas flow meter. The gas permeance (Q, mol m⁻² s⁻¹ Pa⁻¹) was calculated as:

Q = J c_i/(Δp A_GF)

(2)

where J—the flow rate of the mixture of investigated gas and carrier gas, c_i—the concentration of investigated gas in the mixture with carrier gas in the receiver measured using a gas chromatograph, A_GF—the effective area of graphite foil, Δp—the partial pressure drop of the investigated gas across the GF (Δp = 1 atm).

2.2.3. Thermal Properties Measurements

For the thermal stability measurements, GF samples (5 cm × 5 cm × 1.5 cm) with a density of 1 g cm⁻³, having previously been weighed, were placed in an oven heated to 670 °C and kept there for 60 minutes. The samples were then removed from the oven and weighed. The mass loss (Δm₆₇₀, %) was calculated as Δm₆₇₀ = (m₀ − m_t)/m₀ 100%, where m₀ is the initial GF mass, m_t is the GF mass after calcination at 670 °C.

To determine the thermal parameters, the thermogravimetry method was used. The studies were carried out on a NETZSCH STA 449C Jupiter thermal analyzer at three heating rates of 5, 10 and 15 °C min⁻¹, in a flow of dried air with a flow rate of 70 mL min⁻¹ in the temperature range of 40–1000 °C. The obtained experimental data were processed using Netzsch software packages: “Proteus Analysis”, “Peak Separation”, “Thermokinetics”.

At the first stage of data processing, the temperatures of the oxidation beginning of the samples were determined. For this thermal gravimetry curves obtained at the heating rate of 5 °C min⁻¹ were used. The temperature at which the graphite foil begins to oxidize is the temperature at which the sample loses 1% of its mass (T_1%, °C).

In order to determine the kinetic parameters of the oxidation of graphite foil with air oxygen, thermogravimetric analysis was performed at three different heating rates of 5, 10 and 15 °C min⁻¹. In order to prove that the reaction mechanism does not change depending on the heating rate, the degree of conversion α versus the normalized temperature T/T_50% were plotted for each heating rate. Normalization of the temperature scale to the base temperature was carried out at the same degree of conversion α = 0.5 for all compared curves. The fact that these graphs for three different heating rates coincide allows us to say that the oxidation proceeds in the same way, regardless of the heating rate. Plots of α versus T/T_50% were the same for all samples.

Next, the kinetic parameters of the graphite foil oxidation reaction were determined. The dependence of the change in the reaction rate on temperature is described by an equation [38] of

−dα/dt = Ae^−Ea/RTf(α)

(3)

A and E_a are the Arrhenius parameters, A is the preexponential coefficient, E_a is the apparent activation energy (the temperature coefficient of the reaction rate), T is the temperature, f(α) is the function of the degree of conversion of the ongoing process, which characterizes its mechanism. An analysis of the GF oxidation data made it possible to determine the kinetic parameters of oxidation and the function f(α) of this equation, which approximate the oxidation process from a statistical point of view in the best way.

3. Results and Discussion

3.1. The Peculiarities of the Inner Structure of Compressed Exfoliated Graphite with a Respect to Preparation

The interaction between graphite, potassium dichromate and concentrated sulfuric acid in the presence of various amounts of oxidizer K₂Cr₂O₇ leads to the stoichiometric formation of the graphite intercalation compounds with sulfuric acid [C_24n]⁺HSO₄⁻∙2H₂SO₄ (n—stage number). The formation of GIC with a certain stage was confirmed by XRD (Figure 1a) from the calculation of the identity period (I_c) along the trigonal c axis (

Table 1. The values of the identity period of GICs-H₂SO₄ and mass gain of expandable graphite.

n	Ic, nm	Δm, %
II	1.129	26.0
III	1.450	18.9
IV	1.814	6.0

). GIC identity period, which is the width of the repeating fragment of the crystal lattice, decreases with decreasing the GIC stage number, i.e., with the increase in oxidation degree of the graphite matrix and the amount of intercalate in it. The XRD pattern of hexagonal graphite is characterized by the presence of line (00l), l = 2n, while the remaining reflections (00l) are forbidden due to the action of the helical axis 6₃ in a cell with an alternating structure ABABAB. The entire set of peaks (00l) is visible on the XRD pattern of the GICs, and their extinction does not occur due to a change in the symmetry of the space group of the crystal lattice and the appearance of axis 6 in a cell with an alternating structure AαAαAα… (A is a graphite layer, α is an intercalate layer).

The water hydrolysis of GIC leads to the formation of a nonstoichiometric product, expandable graphite. According to XRD analysis, an interlayer distance of expandable graphite is 0.336 nm (Figure 1b). The local overoxidation of the graphite matrix during intercalation and hydrolysis leads to the formation of oxidized graphite with oxygen functional groups sandwiched between crystalline graphite. The presence of oxygen groups was confirmed by FTIR spectroscopy (Figure 1c). These groups are formed only on the surface layers and side edges of the graphite crystallites, and the expandable graphite planar structure is preserved. Mass gain of expandable graphite (Δm, %) increases when decreasing the GIC stage number (Table 1), which is indicative of an increase in the amount of oxidized regions in the expandable graphite matrix.

Thus, in the process of obtaining of expandable graphite, the morphology, microstructure and crystalline structure of the initial graphite undergoes a number of changes (Figure 2). During the intercalation and hydrolysis, significant delamination of particles of expandable graphite based on GIC of stage II occurs, while particles based on GIC of stage IV practically do not laminate and are similar in morphology to the original graphite (Figure 2a,b). The formation of expandable graphite is accompanied by a decrease in the crystallite size along the c axis L_c and the formation of functional oxygen groups on the surface and side grains of the crystallites (Figure 2c,d). An increase in the degree of oxidation of the graphite matrix during the synthesis of II stage GIC leads to the formation of a higher amount of oxygen groups in the structure of expandable graphite (Figure 2e).

With flash heating, the exfoliation of the graphite packs along the c axis occurs, the degree of which increases when decreasing the GIC stage number (Figure 2). According to XRD analysis, all samples of exfoliated graphite EG-n-T and materials compressed from GF-n-T have the graphitic structure with interlayer distance of 0.336 nm (Figure 1d). Exfoliation of expandable graphite packs leads to the formation of many carbon walls in the structure of the EG particle (Figure 2a–c), and the thickness of the walls is comparable to the crystallite size L_c (Figure 2d). Thus, EG exfoliation volume (V_EG, mL g⁻¹) increases when decreasing the GIC stage number and increasing the EG preparation temperature (Figure 3a), which is connected with increasing the pressure of the releasing gases in the graphite matrix. These gases are the products of the decomposition of oxygen groups and remaining intercalate in the expandable graphite. The smaller the GIC stage number, the more oxygen groups are present in the structure of expandable graphite and the higher the amount and pressure of the releasing gases (CO₂, CO, H₂O, O₂).

On the other side, an increase in the EG preparation temperature leads to oxidation of forming exfoliated graphite and a decrease of EG yield (Y_EG, %) (Figure 3b). The higher temperature promotes the rate of oxidation of the graphite matrix during exfoliation. Thus, the EG obtained at 1000 °C is characterized by less yield in comparison with lower temperatures.

Graphite foil compressed from exfoliated graphite consists of the arranged graphite sheets formed by the process of adhesion interaction between graphite packs with the numerous cavities between them. Gas permeability is limited by narrow pores between and inside graphite packs connecting wide pores in the material. The formation of porous GF structure depends on the number and the area of contacts between the graphite packs, which in turn depend on EG exfoliation volume.

3.2. Influence of the GIC Stage Number on the Pore Structure and Gas Permeability of Graphite Foil

The investigation of nitrogen and hydrogen permeance (Q, mol m⁻² s⁻¹ Pa⁻¹) was carried out in a diffusion cell by a differential method. An increase in gas permeance of GF with increasing the GIC stage number is observed for a fixed EG preparation temperature (Figure 4). The most significant increase in gas permeance values with increasing the stage number of initial GIC is for EG preparation temperature of 600 °C. GF-II-600 has gas permeance that is two orders of magnitude less than gas permeance of GF-IV-600 (Figure 4a).

When graphite foil is compressed from EG, an increase in the EG exfoliation volume leads to an increase in the area and the number of contacts between EG particles and an increase in the adhesive interaction between them. That in turns leads to the closing of the macropore space for gas transport and the formation of GF with smaller macropore size. Macropore volume (V_macro, cm³ g⁻¹) of GF-IV-800 is 0.55 cm³ g⁻¹ while GF-II-800 has a lower macropore volume of 0.47 cm³ g⁻¹ (

Table 2. Gas transport and porous characteristics of GF based on II, III, IV stage GIC (EG preparation temperature = 800 °C).

Sample	Q(N2)∙1010, mol m−2∙s−1∙Pa−1	Q(H2)∙1010, mol m−2∙s−1∙Pa−1	α(H2/N2)	w, nm	Vmacro, cm3 g−1	Vmeso, cm3 g−1
GF-IV-800	14.9	58.8	3.9	160	0.55	0.022
GF-III-800	1.7	6.7	4.0	130	0.51	0.052
GF-II-800	0.8	3.5	4.6	100	0.47	0.060

). According to mercury porosimetry results, the characteristic macropore size (w, nm) of GF-n-800 decreases from 160 nm to 100 nm when decreasing GIC stage number (Figure 5). Thus, minimum permeance is reached for GF with less macropore size and macropore volume.

Gas transport in pores of a material can occur by Knudsen diffusion or viscous flow depending on the Knudsen number, which is the ratio of the mean free path of gas molecules to the characteristic width of the pores. In the case of small Knudsen numbers K_n < 0.1, the motion of gas through a porous material is described by a viscous flow. The transitional regime of gas movement through a material that combines viscous and molecular flows is characterized by Knudsen numbers in the range 0.1 < K_n < 10 [39]. In the case when the Knudsen number becomes greater than 10, Knudsen diffusion becomes predominant in gas transfer through a porous material. For the samples GF-n-800 the calculated ideal selectivity of separation of H₂/N₂ α(H₂/N₂) = Q(H₂)/Q(N₂) (Table 2) is greater than the theoretical Knudsen selectivity α(H₂/N₂) = (M_N2/M_H2)^0.5 = 3.7. Generally, ideal selectivity decreases when increasing the stage number, which confirms the increase in the contribution of pores with larger size in gas transport. Probably, gas transport and separation predominantly occur in low sized micro- and mesopores, which connect wider macropores.

3.3. Influence of the GIC Stage Number on the Thermal Properties of Graphite Foil

The lower the stage number of the initial GIC, the greater dispersion of the crystallites of the original graphite and an increase in the defectiveness of the GF structure are achieved. For GF-n-800, the crystallite sizes along c axis L_c decrease from 27 to 22 nm when decreasing in the step number. Thus, the BET surface area and the number of active oxidation centers increase (

Table 3. Structure characteristics and kinetic analysis data of GF samples based on II, III, IV stage GIC (EG preparation temperature = 800 °C).

Sample	Lc, nm	sBET, m2 g−1	Δm670, % h−1	T1%, oC	Ea, kJ mol−1	Kinetic Equation −dα/dt = Ae−Ea/RT(1 − α)nαa
GF-IV-800	27	11.4	18.4	580	156	105.3e−156/RT(1 − α)0.97α0.12
GF-III-800	24	23.6	19.2	563	142	104.6e−142/RT(1 − α)0.96α0.13
GF-II-800	22	26.9	21.0	545	112	103.1e−112/RT(1 − α)0.93α0.25

).

For GF-n-800, a regular decrease in weight loss at 670 °C (Δm₆₇₀) was observed with an increase in the GIC stage number (Table 3). According to thermogravimetric analysis, GF begin to oxidize at a temperature above 500 °C (Figure 6). The temperature at which the graphite foil begins to oxidize is the temperature at which the sample loses 1% of its mass (T_1%, °C). There is a certain dependence of the temperature of the beginning of oxidation on the GIC stage number. An increase in the GIC stage number leads to an increase in the temperature of the start of oxidation (Table 3).

The kinetic parameters of the GF oxidation reaction were determined. The dependence of the change in the reaction rate on temperature is described by an equation of −dα/dt = Ae^−Ea/RTf(α). The oxidation function f(α) of samples GF-n-800 is best described by the Prout–Tompkins equation f(α) = (1 − α)^mα^a, the relative values of a and m determine relative contributions from the acceleratory and decay processes (Table 3) [40]. This equation describes autocatalytic processes proceeding according to the mechanism of chain branched nucleation. In this case, the nuclei are the new active oxidation centers. The chain nature of this process can be explained by the lower content of active centers and their location on the edges of the graphite layers. In this regard, at the end of the oxidation of the outer layers in contact with oxygen, access to a larger amount of new potential oxidation centers is opened.

The calculated values of the apparent activation energy (E_a, kJ mol⁻¹) decreases with decreasing the GIC stage number (Table 3). The calculated activation energy is just apparent due to the fact that in the considered process the course of the reaction depends on the morphology of the sample surface. In addition, in the oxidation of GF with oxygen, the most important stage is the rate of oxygen sorption and desorption of reaction products. These processes affect the value of the apparent activation energy. The GF with more dispersed crystallites with less size (L_c, nm) and more defective structure has a higher surface area (s_BET, m² g⁻¹) available for oxidation (Table 3). The smaller activation energy is in good agreement with these structure parameters and the assumption about the occurrence of oxidation.

3.4. Influence of the EG Preparation Temperature on the Pore Structure and Gas Permeability of Graphite Foil

EG preparation temperature is the second preparation condition of EG, which influences the gas permeance. Gas permeance of GF based on GIC of II, III, IV stages decreases when increasing the EG preparation temperature from 600 to 800 °C and increases when increasing temperature up to 1000 °C (Figure 7).

The increase in GF gas permeance when decreasing the EG preparation temperature from 800 °C to 600 °C is connected with the change in the same structural parameters, which influence the increase in GF permeance when increasing the GIC stage number. The decrease in the number of contacts between graphite packs during the pressing of EG-II-600 leads to the opening of some of the macropores inside of GF and the increase in macropore size and volume. According to mercury porosimetry, the characteristic macropore width of GF-II-T increases when from 100 nm to 130 nm when decreasing the EG preparation temperature from 800 °C to 600 °C (Figure 8). The sample GF-II-600 has higher macropore volume than GF-II-800 (

Table 4. Gas transport and porous characteristics of GF samples based on II stage GIC (EG preparation temperature = 600, 800, 1000 °C).

Sample	Q(N2)∙1010, mol m−2∙s−1∙Pa−1	Q(H2)∙1010, mol m−2∙s−1∙Pa−1	α(H2/N2)	w, nm	Vmacro, cm3 g−1	Vmeso, cm3 g−1
GF-II-600	1.0	6.0	5.9	130	0.52	0.050
GF-II-800	0.8	3.5	4.6	100	0.47	0.060
GF-II-1000	5.8	19.7	3.4	100	0.45	0.067

). Thus, a material with a higher amount of transport pores is formed. This increase in permeance for the samples GF-IV-600 is more significant in comparison to the other samples. GF-IV-600 has a maximum characteristic macropore width of 195 nm. The nitrogen and hydrogen permeances of GF-IV-T increases from 14.9∙10¹⁰ to 39.8∙10¹⁰ mol m⁻²∙s⁻¹∙Pa⁻¹ and from 58.8∙10¹⁰ to 112∙10¹⁰ mol m⁻²∙s⁻¹∙Pa⁻¹, respectively, when decreasing the EG preparation temperature from 800 to 600 °C (Figure 7).

The increase in EG preparation temperature from 800 to 1000 °C leads to an increase in permeance of GF based on GIC of II, III, IV stages (Figure 7) despite the increase in EG exfoliation volume and the number of contacts between graphite packs during the pressing of EG-n-1000. According to the mercury porosimetry data, the macropore volume and characteristic width of macropores of GF-II-T practically does not change when increasing the EG preparation temperature from 800 to 1000 °C (Table 4). The nitrogen adsorption–desorption method shows that these samples have mesopores with a width of less than 50 nm. The BET surface area (s_BET, m² g⁻¹) and the mesopore volume (V_meso, cm³ g⁻¹) of samples GF-n-T increases when increasing the EG preparation temperature up to 1000 °C (Figure 9a,b).

The increase in the GF mesopore’s volume can be related to the oxidation of carbon in the intercrystallite space of exfoliated graphite during the thermal treatment at a higher preparation temperature. Oxidation is related to the change in EG yield, which decreases when increasing the EG preparation temperature (Figure 3b). The oxidation of the carbon matrix leads to the formation of new paths for gas transport in the intercrystallite space, which in turn leads to an increase in the gas permeability of GF-n-1000 pressed from this exfoliated graphite.

The ideal selectivity of separation of H₂/N₂ α(H₂/N₂) also decreases simultaneously with the increase in permeability of these gases (Figure 10). This is connected with an increase in the porosity and the opening of new pores with larger size, which participate in gas transport. Thus, the selectivity for the samples GF-n-1000 becomes less than Knudsen selectivity. Permeability data for the sample GF-IV-600 does not align with this dependence (Figure 10) because of weak contacts between compressed EG particles and, respectively, participation of a large amount of macropores in gas transport.

Thus, there are two competing processes during the formation of exfoliated graphite. One of them is an increase in the EG exfoliation volume with both decreasing the GIC stage number and increasing the EG preparation temperature. The increase in EG exfoliation volume leads to the increase in the number of contacts between EG particles during its pressing into graphite foil and the decrease in GF macropore size and gas permeance, respectively. The second process is the oxidation of the EG matrix when increasing the EG preparation temperature, which leads to the formation of transport mesopores in exfoliated graphite and GF compressed from it. Respectively, graphite foil with minimal permeability is formed from EG obtained at conditions, at which the high EG exfoliation volume and the low oxidation degree of EG matrix are reached. GF-II-800 is characterized by minimum permeability among all samples.

3.5. Influence of the EG Preparation Temperature on the Thermal Properties of Graphite Foil

The oxidation of samples GF-II-T is also described by the Prout–Tompkins equation. A decrease in the size of crystallites L_c and an increase in the specific surface area of GF samples are observed when increasing the EG preparation temperature (

Table 5. Structure characteristics and kinetic analysis data of GF samples based on II stage GIC (EG preparation temperature = 600, 800, 1000 °C).

Sample	Lc, nm	sBET, m2 g−1	Δm670, % h−1	T1%, oC	Ea, kJ mol−1	Kinetic Equation −dα/dt = Ae−Ea/RT(1 − α)nαa
GF-II-600	25	24.5	24.7	542	121	103.5e−121/RT(1 − α)0.79α0.14
GF-II-800	22	26,9	21.5	545	112	103.1e−112/RT(1 − α)0.93α0.25
GF-II-1000	18	31.0	21.4	549	101	103.1e−101/RT(1 − α)0.98α0.34

). Thus, the defectiveness of the GF samples increases, which in turn leads to an increase in the number of active centers and a decrease in the apparent activation energy (Table 5).

However, the dependence of the temperature of the GF oxidation beginning does not correlate with the activation energy (Table 5). The temperature at which the GF oxidation begins, on the contrary, slightly increases with an increase in the EG preparation temperature (Figure 11). This is presumably due to the presence of residual low-sized carbon fragments in the structure of GF, which are present in a significant amount in GF-II-600 and are practically absent in GF-II-1000. The higher EG preparation temperature leads to the oxidation of these carbon fragments at the stage of exfoliation and the formation of transport pores in the structure of the obtained GF [3]. On the contrary, these fragments were preserved in the composition of EG obtained at lower temperature and block the transport pores in the obtained GF. These carbon fragments in GF-II-600 begin to decompose at a lower temperature than the whole graphite matrix, and the weight loss of the sample increases at a lower temperature (Table 5).

4. Conclusions

A systematic study of the physicochemical properties of graphite foil based on exfoliated graphite showed that the GF gas transport properties and thermal stability are determined by GF structural features. They in turn depend on the conditions of GF preparation: the GIC stage number and the EG preparation temperature. The structure of the initial graphite undergoes significant changes during the subsequent stages of intercalation and thermal expansion. There is a noticeable increase in EG exfoliation volume and a decrease in the crystallite size along the c axis with a decrease in the GIC stage number and with an increase in the EG preparation temperature. Thus, graphite foil pressed from EG based on GIC of a low stage has a lower macropore volume and size, which is due to a higher number of contacts between EG particles. On the other side, the increase in EG preparation temperature from 800 to 1000 °C does not change the macroporosity of compressed GF but leads to an increase in the mesopore volume GF and, respectively, the formation of new transport pores. Low macropore volume and size lead to low hydrogen and nitrogen permeability of GF. However, an increase in mesoporosity of the GF results in an increase in GF permeability.

GF based on GIC of a lower stage has a larger specific surface area and, consequently, a larger number of active oxidation sites. All samples have a start oxidation temperature above 500 °C. Both the temperature of the start of oxidation and the values of the apparent activation energy of graphite foil decrease when decreasing the stage number of the initial GIC, which is due to an increase in the number of active oxidation sites. With an increase in the EG preparation temperature, the apparent activation energy also decreases, which is due to an increase in the defectiveness of the graphite matrix. The oxidation process of all GF samples is best described by the Prout–Tompkins equation, which describes autocatalytic processes. Thus, the optimum preparation temperature of EG and GF is 800 °C, at which the minimum H₂ and N₂ permeability is observed for each GIC stage number. On the other hand, GF based on GIC of stage II has lower thermal stability. Thus, for relatively low-temperature applications such as gas condensation and separation, using overheated steam GF based on GIC of stage II can be used. For high-temperature processes such as oil refining, a GF based on GIC of a higher stage with high temperature resistance should be selected.