Effect of Tensile Strength on the Microstructure of Graphite Impregnated with Salt Revealed by In Situ Synchrotron-Based Two-Dimensional X-ray Diffraction.

Owing to the inhomogeneous distribution of FLiNaK salt impregnated into graphite which is observed by scanning electron microscopy and an element probe micro-analyzer, a map scan of in situ real-time tensile synchrotron-based two-dimensional X-ray diffraction (2D-XRD) at several fixed external forces was implemented to reveal the local microstructure evolution of graphite and FLiNaK salt. Notably, a stress concentration area (SCA), that is, the main interaction area between graphite and salt, was found and then transformed from one region to another region because of the unbalanced squeeze interaction between graphite and FLiNaK salt with the increase of external force. During the external stress load process, a smaller grain size, poorer crystallinity of graphite and a larger grain size, better crystallinity of FLiNaK salt appear in the SCA; meanwhile, the changes of crystallographic preferred orientation of FLiNaK salt domains in SCA imply that the external load force makes better the ordered stacking of the larger crystal grains of the FLiNaK salt impregnated into graphite. Most importantly, we have found for the first time that the fracture position of graphite impregnated with FLiNaK salt always occurs near the SCA rather than at a fixed region under the external stress load. Thus, the present study not only helps to reveal the interaction mechanism between graphite and FLiNaK salt under the external stress load but also contributes to accurately predict and analyze the stress state of components, which would have an effective impact on the design of a molten salt reactor and the reliability of the component safety assessment.

Introduction

A molten salt reactor (MSR) is one of the six candidate reactors in the fourth generation of the fission nuclear energy system.[1,2] It is characterized by molten fluoride salt as the carrier of coolants or fuels.[3] In MSRs, graphite material is used as the neutron moderator and reflector because of a series of advantages, such as a high neutron scattering cross section, low neutron absorption cross section, high density, and good thermal conductivity, and so forth.[3,4] Graphite is also used as the core of structural material in the MSR, which needs to ensure the structural integrity of the internal components of the reactor. One of the key factors affecting the structural integrity of graphite components is the mechanical properties of graphite materials,[5] such as the bearing between graphite components, the asymmetry of component structures, and the concentration of load-bearing stress caused by unevenness.[6,7] It is well known that the mechanical properties of graphite are affected by the internal strain within its microstructure. For example, the permanent deformation encountered by nonirradiated graphite during the tensile process occurs on the length scale of the microstructure because of the nonuniform microstructure.[8,9] The change in pore volume caused by mechanical stress under the tension process plays an important role in the reduction of Young’s modulus.[10] It has also been found in the neutron and X-ray studies of polycrystalline graphite that the application of tensile strain reduces the bulk elasticity modulus, and in the case of tensile, the relationship between the elastic strain and the applied strain in the graphite crystal is nonlinear.[8] After the strained graphite is annealed, the residual strain is alleviated and the stress–strain characteristics are also restored.[11]

Furthermore, graphite requires direct contact with molten salt in the MSR.[12] Owing to the complex porous structure of graphite, molten salt easily penetrates into the pores of graphite under a certain pressure, which would affect the mechanical properties of graphite and therefore shorten the service life of graphite material, resulting in a huge impact on the operation of the MSR.[13] For example, Zhang et al. found that the compressive strength tests show that the graphite sample impregnated with FLiNaK salt shows a longitudinal splitting fracture rather than the shear fracture of virgin graphite, and the corresponding failure mechanism change is suggested to be an extra stress generated in graphite impregnated with FLiNaK salt.[13] Therefore, it is indispensable to study the molten salt impregnation behavior of graphite in the MSR. In our previous reports, we have found that the normal temperature compressive strength of graphite increased significantly after FLiBe molten salt impregnation, and with the increase of impregnation, the compressive strength of graphite at room temperature increases, which is mainly attributed to the fact that the solid FLiBe salt impregnated into the graphite increases the load-bearing area of the graphite material and reduces the stress under a certain external force.[14,15] It is also found by Qi et al. that the d002 spacing of a graphite layer decreases with increase of the amount of molten salt impregnation, indicating that the microstructure of graphite and molten salt has a certain interaction, which has an effective impact on the macroscopic mechanical properties of graphite.[16] It is widely known that the interaction between graphite and molten salt leads to a change in the pore structure of the graphite, which is bound to affect the internal stress distribution of the graphite. Meanwhile, the internal stress field distribution of the graphite material will further evolve with a change of the external stress load. With increase of the external stress load, defects and damages in the graphite internal microstructure are further accumulated, eventually leading to material fracture and component failure.[17,18] Therefore, it is of great significance to study the fracture behavior of graphite impregnated with FLiNaK salt under external tensile force. However, there have been few reports on in situ studies on the tensile fracture of graphite impregnated with molten salt.

Herein, in situ tensile synchrotron-based two-dimensional X-ray diffraction (2D-XRD) was used to characterize the tensile fracture behavior of graphite impregnated with FLiNaK salt, which can reveal the real-time microstructure interaction between graphite material and the molten salt. Raman spectroscopy, scanning electron microscopy (SEM), X-ray energy-dispersive spectroscopy (EDS), and an electron probe micro-analyzer (EPMA) were further used to analyze the defects, morphology, and element distribution of graphite impregnated with FLiNaK salt before and after the fracture. Owing to the inhomogeneous distribution of FLiNaK salt impregnated into graphite bulk, a map scan (11 points × 3 lines with a beam size of 0.2 × 0.2 mm2) of in situ synchrotron-based 2D-XRD at several fixed forces was carried out to reveal the local microstructure evolution and their corresponding strain distribution. Notably, it is found that during the application of the external load, the strain concentration area (SCA), that is, the main interaction area between graphite and salt was found, which would lead to the fracture of graphite. With increase of external force, the SCA has been transformed from one region to another region because of the unbalanced squeeze interaction between graphite and salt resulting from the inhomogeneous distribution of FLiNaK salt impregnated into graphite bulk. Thus, all of the results will contribute to understand the interaction mechanism between graphite and FLiNaK salt and help to explain the change of mechanical properties of graphite impregnated with molten salt, which are conducive to the fabrication of high-performance graphite and safe operation of the MSR.

Results and Discussion

Figure a shows the obvious pore characteristics of the NBG-18 graphite with pore size from several microns up to subhundred microns before FLiNak salt impregnation, mainly resulting from the aggregate particles and binders of NBG-18 graphite during the fabrication process.[19,20]Figure b shows the EDS analysis of the C element of Figure a, which further confirms the pores presented in graphite. These pores provide a channel through which the molten salt can penetrate from the surface into the microstructure. It is well known that graphite contains two types of pores: closed pores and open pores.[21] Salt occupies open pores, and the distribution of salt in graphite represents the distribution of open pores. Figure c shows a morphology image of the graphite surface after FLiNaK salt impregnation, and Figure d shows an EDS analysis of the element F of Figure c, which shows an observable inhomogeneous distribution of the salt in graphite. Figure S1a shows the morphology of the fracture surface of graphite. Studies have demonstrated that the pores inside the graphite are a network structure.[22] It can be predicted that the molten salt impregnated into the graphite will also have a network structure after the molten salt impregnation experiment. As can be seen from the enlarged view of the salt of Figure S1b that some of the salt exhibits a network structure, which is well consistent with the above guess. The distribution of salt on a large area of the fracture surface of the graphite sample impregnated with FLiNaK salt after the tensile fracture was analyzed using an EPMA, as displayed in Figure e. The distribution of F (Figure f), Na (Figure S2a), and K (Figure S2b) is compared with the second electron image (Figure e). As shown in Figures f and S2a,b, the distribution of F, Na, and K is superposed and their distributions are also the same as the white-colored area in the second electron image shown in Figure e. It is found that the distribution of surface salt of the fracture surface is also uneven. Owing to the uneven distribution of FLiNaK salt impregnated into the graphite bulk, it is absolutely necessary that a map scan of in situ 2D-XRD should carry out at different areas of graphite to reveal the local microstructure evolution during the process of applying external stress.

Figure 1

(a,b) SEM and EDS images of the graphite sample without FLiNaK salt impregnation. (c,d) SEM and EDS images of the graphite sample impregnated with FLiNaK salt. (e,f) Second electron image and EPMA analysis about the F element of the fracture surface of the graphite sample impregnated with FLiNaK salt.

Figure a shows the in situ 2D-XRD experimental setup based on the BL14B1 diffraction station of Shanghai Synchrotron Radiation Facility (SSRF) with X-ray beam size 200 × 200 μm2 at 18 keV. In comparison to the conventional X-ray diffraction facilities, synchrotron-based X-ray diffraction could achieve a higher-quality powder diffraction patterns in terms of the peak profile shape and the full width at half-maximum (FWHM) resolution,[23] which enables better resolving capability as well as fitting results for both qualitative and quantitative measurements. A series of in situ tensile synchrotron-based 2D-XRD experiments were carried out on the NBG-18 graphite sample after FLiNaK salt impregnation with a photo of the experimental setup as shown in Figure b. The inset in Figure b is a photograph of the test graphite sample. Figure c clearly shows a schematic representation of the test graphite sample. Prior to testing, a wire cutter was utilized to cut a notch that is approximately one-third the length of the sample. The notch has a diameter of 0.35 mm. A total of three lines are measured, each of which measures 11 positions. Position 6 of each line is approximately at the center of the notch. The X-ray beam size of the SSRF BL14B1 line station is 200 × 200 μm2, and the step of each movement is 100 μm. The square with a side length of 200 μm represents the beam, and the red solid circle represents the measured position, as displayed in Figure c. Figure d shows the relationship between tensile force and displacement, which obviously demonstrates that before the external stress load is 5 N, the displacement increases faster with the increase of the external stress load and after the external stress load is 5 N, the displacement slowly increases with the increase of the external stress load. In situ real-time 2D-XRD diffraction technology is further adopted based on the fast 2D area detection X-ray source of high-brightness synchrotron, which can provide more colorful crystal structure information. Figure e–g shows the 2D-XRD profiles of graphite impregnated with FLiNaK salt at tensile forces 0, 15, and 27 N. The narrow and spotty scattered rings at q ≈ 26.9 nm–1 in Figure e–g are from the LiF(111)/NaF(200) diffraction peak of FLiNaK salt, which indicates that the molten salt with good crystallization has impregnated into the graphite sample after the FLiNaK salt impregnation experiment. It can be seen from Figure e–g that the diffraction ring has undergone significant changes during the external stress load process. This mainly reflects the interaction between the graphite layer and the molten salt during the external stress load process.

Figure 2

(a) X-ray diffraction (XRD) experimental setup. (b) Photograph of the setup for the in situ stretching synchrotron-based XRD experiments, where the inset shows graphite mounted for the measurements. (c) Schematic diagram of the test sample and test location. (d) Curve of displacement of the graphite sample with tensile force. (e–g)In situ synchrotron-based 2D-XRD patterns of the graphite sample impregnated with FLiNaK salt during the stretching process.

Figure shows a series of mapping distribution pictures of the layer spacing d002, FWHM, and intensity of the (002) diffraction peak of graphite impregnated with FLiNaK salt in the area 1.2 × 0.4 mm2 centered on the notch marked in Figure c, which illustrates that all of the values in the map present an obvious fluctuation due to the inhomogeneous distribution of FLiNaK salt impregnated into the graphite bulk. The external load stress values tested in different positions are 0, 15, and 27 N, respectively. Figure a shows that the layer spacing d002 of graphite without external stress load is relatively small near position 5 in the map because of the squeeze between the salt and graphite layers,[24,25] implying that this area shows an obvious interaction between graphite layers and salt domains, which can be called the stress concentration area (SCA). Figure b,c shows that d002 in the map represents a slight decrease under 15 and 27 N load, indicating that the squeeze between the salt and graphite layers could be improved by external stress load. Notably, the SCA changed from the position 5 (0 N) to position 8 (27 N), which demonstrated that the graphite layer spacing in the interaction areas between the salt and graphite was different after the addition of the external load stress because of the inhomogeneous distribution of FLiNaK salt impregnated into the graphite bulk. Notably, we have found that the fracture area of graphite impregnated with FLiNaK salt is near position 8 based on a series of tensile tests. Figure d–f shows the variation of the FWHM of the (002) diffraction peak in different areas, which indicates that the different extrusions of salt in the map can result in different size increases of graphite grains. Figure d shows that FWHM of the (002) diffraction peak of graphite impregnated with FLiNaK salt near SCA (position 5) is relatively larger than that of other areas when no external stress was applied, implying that a smaller size of graphite grain appeared in the SCA. After the external load was applied, the FWHM of position 8 (27 N) is larger than that of other areas, which is well consistent with the change of SCA discussed above, which further indicates that the external load stress reduces the grain size of graphite impregnated with FLiNaK salt. Figure g–i reflects the variation of the intensity of the (002) diffraction peak in different areas, which indicates that the different extrusions of salt in the map have an influence on the crystallinity of graphite impregnated with FLiNaK salt. Figure g shows that the intensity of the (002) diffraction peak of graphite impregnated with FLiNaK salt in SCA (position 5) is relatively smaller than other areas when no external stress was applied, implying that the crystallinity of the graphite impregnated with FLiNaK salt becomes worse in the SCA. After the external load was applied, the intensity of position 8 (27 N) is smaller than that of other areas, which indicates that the external load stress reduces the crystallinity of the graphite impregnated with FLiNaK salt.

Figure 3

Diffraction parameter mapping of the (002) peak of graphite impregnated with FLiNaK salt. (a–c), (d–f), and (g–i) Images based on the values of d002, FWHM, and the intensity, respectively.

In order to understand the interaction between the microstructure of the graphite and the salt, the crystallinity of salt in the SCA was also analyzed. A series of LiF(111)/NaF(200) diffraction peaks of FLiNaK salt were further analyzed. Figure shows a range of mapping distribution pictures of the layer spacing dLiF111/NaF200, FWHM, and intensity of the LiF salt (111)/NaF salt (200) diffraction peak in the same region 1.2 × 0.4 mm2, which demonstrates that all of the values in the map show a distinct fluctuation, indicating an obvious different crystallinity of salt impregnated into graphite bulk because of the inhomogeneous distribution of FLiNaK salt. Figure a shows that the layer spacing dLiF111/NaF200 of the salt is relatively small in the SCA (position 5) in the map when the external load is 0 N, which is attributed to the squeeze between graphite and the salt. After the addition of the external load stress, Figure b,c shows that dLiF111/NaF200 in the map in the SCA (position 8) is smaller than that in other areas, which demonstrates that the main interaction areas between graphite and the salt have been changed from position 5 to 8 because of the unbalanced stress response of graphite and salt. Figure d–f shows the FWHM changes in the LiF(111)/NaF(200) diffraction peak of the LiF/NaF salt at different areas in the map, which indicates that the different extrusions of graphite can result in different size increases of FLiNaK salt domains. As shown in Figure d, the FWHM of the (111)/(200) diffraction peak of LiF/NaF salt in the SCA (position 5) is relatively small when no external stress was applied (0 N), implying a larger size of FLiNaK salt grains in the SCA. After the external load was applied, the FWHM of position 8 (27 N) is smaller than that of other areas, indicating that the external load stress has increased the grain size of FLiNaK salt. Figure g–i reflects the variation of the (111)/(200) diffraction peak intensity of LiF/NaF salt in different areas, which clearly demonstrates that different extrusions of salt in the map have also influenced the crystallinity of FLiNaK salt. Figure g shows that the intensity of the (111)/(200) diffraction peak of LiF/NaF salt near SCA (position 5) is larger than that in other areas when no external stress was applied, implying that the crystallinity of the FLiNaK salt is improved in the SCA. After applying the external load, the intensity of the (111)/(200) diffraction peak at the position 8 becomes larger than that in the other regions, which further confirms that the SCA has been changed from position 5 to position 8 because of the application of external load stress, improving the crystallinity of the FLiNaK salt with different degrees because of the inhomogeneous distribution of FLiNaK salt impregnated into graphite.

Figure 4

Diffraction parameter mapping of LiF/NaF salt. (a–c), (d–f), and (g–i) Images based on the values of dLiF111/NaF200, FWHM, and the intensity, respectively.

It is widely known that the crystallographic orientations of different structural domains in all directions can be examined in detail by radially integrating the corresponding scattered ring.[26] Subsequently, the scattering rings corresponding to the typical (002) crystalline plane at q ≈ 18.7 nm–1 of graphite and the (111)/(200) crystalline plane at q ≈ 26.9 nm–1 of the LiF/NaF salt in different areas were radially integrated and plotted as functions of the azimuth angle,[26,27] as displayed in Figure a,b. Figure a demonstrates clearly that the azimuth orientation of the preferential out-of-plane orientation of the graphite (002) plane is different in the SCA because of the different extrusion effects between the salt and graphite caused by the inhomogeneous distribution of salt. Although the integrated intensity changes of graphite (002) peaks at position 5 and position 8 are not obvious, the transitions of the orientation azimuth degree along the out-of-plane direction show a reverse shift under external load stress, which demonstrates a different texture evolution of graphite domains because of the synergism from salt and external stress. Obviously, as shown in Figure b, the azimuth orientation of the preferential out-of-plane orientation of the LiF/NaF salt (111)/(200) plane in the FLiNaK salt shows a series of sharp peaks and a significant change in intensity at position 5 and position 8 under external load, indicating that the crystallographic orientation of FLiNaK salt impregnated into graphite becomes much more orderly. However, the higher intensity and the narrower peaks of the (111)/(200) salt diffraction peak in position 5 are obvious than the position 8, implying that the crystallinity and order stacking of the FLiNaK salt are better in position 5, which indicates that graphite impregnated with the FLiNaK salt is not easily fractured at this position. These above changes can clearly illustrate a model diagram of an ordered stacking shown in Figure c, which illustrated that the ordered stacking of graphite and FLiNaK salt in a certain azimuth will become more orderly with multiorientation induced by the external load stress. The red squares represent the salt domains, and the lines and curves represent the graphite domains.

Figure 5

(a) Corresponding radially integrated intensity plots along the ring of q ≈ 18.7 nm–1, assigned to the (002) plane of graphite impregnated with FLiNaK salt. (b) Corresponding radially integrated intensity plots along the ring of q ≈ 26.9 nm–1, assigned to the LiF(111)/NaF(200) planes of FLiNaK salt. (c) Model for ordered accumulation of salt crystallites under external stress load.

In summary, these above results observed by in situ synchrotron-based 2D-XRD illustrate clearly the evolution of the SCA under external load stress based on the crystalline changes of graphite and salt, which contribute to the understanding of the reason the fracture broken position of graphite impregnated with salt easily occurs near the SCA, where a smaller grain size, poorer crystallinity of graphite and a larger grain size, good crystallinity of FLiNaK salt during the external load process appear.

In order to further reveal the structure evolution of graphite impregnated with FLiNaK salt induced by each external force used in the present study, the in situ 2D-XRD patterns collected at the center of the map during the whole stretching process are further analyzed as shown in Figure S3. Figure a shows a series of the one-dimensional X-ray diffraction (1D-XRD) spectrum of the (002) diffraction peak of graphite impregnated with FLiNaK salt during the external load process, which is integrated from 2D-XRD patterns as shown in Figure S3. As shown in Figure b, the (002) peak of graphite impregnated with FLiNaK salt presents an obvious shift to the higher 2θ diffraction angle from 11.762° to 11.777° (meanwhile the (004) peak, a senior peak, shows a similar shift trend) and a gradual decrease in the layer spacing d002 during the external load process. It was also observed that the layer spacing d002 has a mutation during the tensile force 3 N, and d002 remains almost unchanged during the subsequent tensile process. Figure c shows the change in FWHM and the intensity of the (002) diffraction peak of graphite impregnated with FLiNaK salt under external stress load. It is clear that the FWHM of the (002) diffraction peak of graphite impregnated with FLiNaK salt first increases from 0.173° (0 N) to 0.179° (5 N), then falls, and finally remains almost unchanged from 20 to 42 N. The intensity ratio of the (002) diffraction peak graphite impregnated with FLiNaK salt first decreased from ∼0.50 to ∼0.13, then increased to ∼1, and finally remained unchanged during the subsequent tensile process. In addition, if we continue to increase the external tensile force from 42 N up to above the threshold tensile, the graphite impregnated with FLiNaK salt in the present study becomes a fracture. Figure d shows a series of the one-dimensional X-ray diffraction (1D-XRD) spectrum of the LiF(111)/NaF(200) diffraction peak of FLiNaK salt during the external load process. As shown in Figure e, the LiF(111)/NaF(200) diffraction peak of FLiNaK salt presents an obvious shift to the higher 2θ diffraction angle from 16.943° to 17.0° and a gradual decrease in dLiF111/NaF200 spacing during the external load process. Figure f shows the relationship of FWHM, intensity, and external load force. With increase of the external load, the intensity of the LiF/NaF salt gradually decreases, and the FWHM gradually increases, indicating the poorer crystallinity of FLiNaK salt during the external load process. Thus, in the present study, the fracture position of graphite impregnated with FLiNaK salt under the external stress load occurs at the SCA rather than at the center of the map as shown in Figure c, which might be attributed to the different residual elastic strain distributions in graphite, resulting from the unexpected pores and inhomogeneous distribution of FLiNaK salt impregnated into graphite. Notably, at the center of the map, the evolutions of the d002 spacing, FWHM, and intensity of the (002) diffraction peak of graphite impregnated with FLiNaK salt during the tensile process indicate that the external load has increased the crystallinity and decreases the layer spacing of graphite impregnated with FLiNaK salt via affecting the release process of graphite residual elastic strains.[8,14,24,25]

Figure 6

In situ synchrotron-based 1D-XRD patterns of the graphite sample impregnated with FLiNaK salt, (002) diffraction peak (a), diffraction angle and layer spacing d002 (b), and the FWHM and peak intensity of the (002) peak (c) as a function of external load force; LiF(111)/NaF(200) diffraction peaks (d), LiF(111)/NaF(200) diffraction angle and dsalt (e), and the FWHM and peak intensity of LiF(111)/NaF(200) diffraction peaks (f) as a function of external load force; and surface Raman contrast of the graphite sample impregnated with FLiNaK salt near the pore position when (g) the external stress load is 0 N and (h) after tensile fracture.

Moreover, Raman spectrum analysis was further implemented to confirm the external load which improves the crystallinity of graphite after FLiNaK salt impregnation, which is one of the most effective tools widely used to characterize carbon materials and can provide effective information on basic microstructures such as in-plane grain size,[28,29] defects,[30] and disorder of carbon materials.[31]Figure g,h shows the Raman spectra of graphite impregnated with FLiNaK salt near the pores before tensile force (0 N) and after the tensile fracture, respectively. The simultaneous presence of the D band, the D1 band, the G band, and the D′ band in the Raman spectrum indicates the presence of defects in graphite. The integrated intensity ratio (R = ID/IG) of the D and G bands is related to the degree of disorder of the carbon materials. When graphite was not subjected to the external load and tensile fracture, the values of R are 0.52 and 0.25, respectively, indicating that the degree of disorder decreased, the defects became less, and the crystallinity became better of NBG-18 graphite after during the external tensile load process. The analysis of the Raman spectrum confirms the above results by XRD.

Conclusions

In conclusion, the real-time interaction of molten salt and graphite during the external tensile load process was investigated by in situ tensile synchrotron-based 2D-XRD. A map scan of synchrotron-based XRD revealed that the SCA of graphite impregnated with the FLiNaK salt appeared because of the inhomogeneous distribution of FLiNaK salt impregnated into graphite bulk when no external load is applied. With increase of external force, the SCA has been transformed from one region to another because of the unbalanced squeeze interaction between graphite and salt. During the application of the external load, the graphite would easily fracture in the SCA, where a smaller grain size, poorer crystallinity of graphite and a larger grain size, better crystallinity of FLiNaK salt appear. Meanwhile, the crystal orientation of the FLiNaK salt indicates that the external load force makes the ordered stacking of the crystal grains of the FLiNaK salt impregnated into the graphite better. However, the ordered stacking of the crystal grains of the FLiNaK salt near position 5 is better than that near position 8 during the application of the external load. Meanwhile, crystal orientation analysis of graphite and salt further indicated that the larger grain size of FLiNaK salt would lead to the fracture of graphite. These findings will help reveal the interaction between the graphite microstructure and the molten salt and be conducive to explain the change of mechanical properties of graphite after molten salt impregnation, which will promote the fabrication of high-performance graphite and safe operation of the MSR.

Experimental Section

Sample Preparation

The research materials in this experiment are the medium-grained grade NBG-18 from SGL Carbon Group, Germany. The size of graphite impregnated with the FLiNaK salt is 20 × 2.5 × 2 mm3. First, the graphite sample needs to be ground and polished. Before impregnation, these were marked by a laser marking machine and then cleaned with acetone, ethyl alcohol, and deionized water as well as dried in vacuum for 2 h at 120 °C to remove the absorbed water. The molten salt was used as a eutectic mixture of LiF, NaF, and KF (46.5 mol % LiF/11.5 mol % NaF/42 mol % KF).[32,33] The melting point of FLiNaK salt was estimated to be 454.0 ± 0.2 °C. Before being impregnated into the molten salt, the graphite samples were degassed using a device. The autoclave was charged with argon gas, and the pressure was kept at 2 atm. The graphite samples were removed from the molten salt after 20 h of impregnation. The pressure in the pressure vessel was maintained until the pressure vessel had cooled to room temperature. The weight of the graphite samples before and after the impregnation test was measured using an analytical balance, which was located in the glovebox. The weight gain of graphite samples is calculated to be (8.59 ± 0.2) wt % by weighing the mass change of graphite before and after the FLiNaK salt impregnation experiment.

Characterizations

SEM observations were carried out using a field-emission scanning electron microscope (LEO 1530 VP). In situ tensile synchrotron-based X-ray diffraction (XRD) was performed at the BL14B1[34] beamline in Shanghai Synchrotron Radiation Facility (SSRF) at a wavelength of 0.06887 nm.[35] The in situ tensile XRD experimental value was measured by the transmission mode, and the graphite NBG-18 was adhered to the already fabricated fixture model and placed at the vertical of the X-ray incidence direction. Two-dimensional XRD patterns were acquired by a MarCCD at a distance of ∼278.65 mm vertically from the sample with an exposure time of 20 s. The XRD patterns were analyzed using FIT2D software and displayed in scattering vector q (q = 4π sin θ/λ, where θ is half of the diffraction angle and λ is the wavelength of incident X-ray) coordinates. The morphology and element distribution of the graphite impregnated with FLiNaK salt were obtained using a LEO 1530VP SEM system and X-ray EDS. Salt distribution after the fracture of the graphite sample impregnated with FLiNaK salt was measured using an EPMA (EPMA-1720). The tensile fracture surface of the graphite sample impregnated with FLiNaK salt was cut by a wire saw and scraped with a stainless steel knife to produce a smooth and fresh surface for EPMA observation. Defect changes of graphite impregnated with FLiNaK salt before and after the tensile fracture were measured using a Raman spectrometer with a 473 nm excitation laser. Calibration of the spectrometer was undertaken with silicon before the measurements. A full spectrum from 200 to 2000 cm–1 was acquired during each measurement with an acquisition time of 10 s.