From Micropores to Ultra-micropores inside Hard Carbon: Toward Enhanced Capacity in Room-/Low-Temperature Sodium-Ion Storage

Highlights Hard-carbon anode dominated with ultra-micropores (< 0.5 nm) was synthesized for sodium-ion batteries via a molten diffusion–carbonization method. The ultra-micropores dominated carbon anode displays an enhanced capacity, which originates from the extra sodium-ion storage sites of the designed ultra-micropores. The thick electrode (~ 19 mg cm−2) with a high areal capacity of 6.14 mAh cm−2 displays an ultrahigh cycling stability and an outstanding low-temperature performance. Abstract Pore structure of hard carbon has a fundamental influence on the electrochemical properties in sodium-ion batteries (SIBs). Ultra-micropores (< 0.5 nm) of hard carbon can function as ionic sieves to reduce the diffusion of slovated Na+ but allow the entrance of naked Na+ into the pores, which can reduce the interficial contact between the electrolyte and the inner pores without sacrificing the fast diffusion kinetics. Herein, a molten diffusion–carbonization method is proposed to transform the micropores (> 1 nm) inside carbon into ultra-micropores (< 0.5 nm). Consequently, the designed carbon anode displays an enhanced capacity of 346 mAh g−1 at 30 mA g−1 with a high ICE value of ~ 80.6% and most of the capacity (~ 90%) is below 1 V. Moreover, the high-loading electrode (~ 19 mg cm−2) exhibits a good temperature endurance with a high areal capacity of 6.14 mAh cm−2 at 25 °C and 5.32 mAh cm−2 at − 20 °C. Based on the in situ X-ray diffraction and ex situ solid-state nuclear magnetic resonance results, the designed ultra-micropores provide the extra Na+ storage sites, which mainly contributes to the enhanced capacity. This proposed strategy shows a good potential for the development of high-performance SIBs. Supplementary Information The online version of this article (10.1007/s40820-020-00587-y) contains supplementary material, which is available to authorized users.


S1 Supplementary Notes
As shown in Fig. S1, the morphology of GC is obviously different from that of the AC (Fig.  1b). The GC particles are in a rod-like morphology and composed of hierarchical flakes while the AC particles are in a dense chunk-like morphology.
As shown in Fig. S2, transmission electron microscopy (TEM) images of the designed carbon, ACGC900, reveal a disordered structure without obvious porosity, which is different from the microporous structure of AC (Fig. 1d). Figure S3 shows the exploration about the effect of the carbonization on the structure of AC. As shown in Fig. S3 and Table S2, the pore structure of AC maintains well after carbonization under 1050 ℃.
As displayed in Fig. S4 and Table S3, the SBET of ACGC900 calculated from CO2 adsorption measurement is ~ 196.6 m 2 g -1 , much higher than that calculated from N2 adsorption/desorption. Moreover, the pore size of ACGC900 is mainly distributed at the range of 0.3-0.5 nm, along with a pore volume of ~ 0.365 cm 3 g -1 .
S2/S20 Figure S5 displays the method of calculating R value from the ratio of the (002) Bragg peak intensity to the background signal. R value (R = B/A) can determine the graphitization degree of the samples. A lower R value suggests a lower degree of graphitization or less stacked graphene layers.
As displayed in the inset table of Fig. S6, the ID/IG value of ACGC900 is lower than that of AC, indicating a higher graphitization degree for ACGC900. Figure S7 displays the first five consecutive CVs of AC, GC, and ACGC900 in a voltage range of 0.001-3.0 V vs Na/Na + at a scan rate of 0.1 mV s -1 . For ACGC900 electrode, an apparent plateau appears at ~ 0.1 V along with the sloping region above 0.1 V in Fig. 2a. Accordingly, a couple of redox peaks at ~ 0.1 V are also present in the cyclic voltammetry (CV) curves (Figs. 2b and Fig. S7). Figure S8 shows that the Na + storage behavior in the carbonized AC electrode keeps the same with that in the pristine AC electrode, demonstrating the significant role of the moltendiffusion process. Figure S9 shows the galvanostatic discharge/charge curves at 50 mA g -1 and CVs in a voltage range of 0.001-3.0 V vs Na/Na + under 0.1 mV s -1 of AC, GC, ACGC900, and AC+GC. For ACGC900 electrode, an apparent plateau appears at ~ 0.1 V along with the sloping region above 0.1 V in Fig. 2A. Accordingly, a couple of redox peaks at ~ 0.1 V are also present in the cyclic voltammetry (CV) curves (Figs. 2b and S7). As shown in Fig. S9, nearly no redox peaks can be observed in the CV curves of the AC+GC electrode, suggesting the significance of the molten-diffusion process.
As shown in Fig. S10, a capacity of more than 100 mAh g -1 can be obtained for the ACGC900 electrode at 2000 mA g -1 while nearly no capacity for the AC electrode, further demonstrating the advantage of the designed structure after the molten diffusion-carbonization, including the reduced interfacial contact between the electrode and electrolyte and the increased degree of graphitization.
As revealed by the SEM images in Fig. S11, after the molten diffusion-carbonization process, the morphology of the micron-sized AC particles is maintained well without obvious residual of PTCDA-derived carbon.
As displayed in Fig. S12, the hysteresis between the adsorption and desorption branches of the isotherm indicates the existence of restricted pore. As shown in Table S1 and Table S3, the SBET of ACGCx decreases with the increasing temperature.
As shown in Fig. S13, the size and SBET of micropores decrease with the increasing carbonization temperature, indicating decreased aperture size and pore inner diameter.
As shown in the Fig. S14 and its inset table, the interlayer spacing decrease from 0.362 nm to 0.350 nm with the increasing carbonization temperature.
As shwon in Fig. 3b, Figs. S14-S16 and Table S4, the sloping capacity contribution decreases linearly with the increasing R value of ACGCx, where a lower R value suggests a lower degree of graphitization or more defect sites. Besides, the initial coulombic efficiency (ICE) value of various samples ranges from 68.1 to 87.9% (Fig. S16).
As shown in Figs. S17 and S18, the decreasing sloping capacity with increasing temperature could be attributed to the reduced population of defect sites, which can be further demonstrated by the relationship between the sloping capacity contribution and the ID/IG ratio obtained from Raman spectra. Figure S19 shows that the initial coulombic efficiency (ICE) value increases linearly with the decreasing SBET from N2 adsorption/desorption measurement (Table S5), indicating reduced parasitic reactions towards electrolyte.
As shown in Fig. S20, the ACGC1050 electrode displays the best rate performance and cycling stability. Specifically, the ACGC1050 electrode can reach ~118 mAh g -1 even at 2000 mA g -1 . Moreover, ~97.3% of the initial capacity can be maintained after 200 cycles at 50 mA g -1 .
In Fig. S21, the hysteresis between the adsorption and desorption branches of the isotherm indicates the existence of restricted pores. Besides, the SBET of ACGCx decreases with increasing temperature.
As shown in Figs. S22 and S23, CMK8GC electrode derived from the mesopore-dominated carbon host (CMK-8) displays no plateau capacity during discharge-charge process. This can be explained that the mesopore inside CMK-8 (cubic Ia3d, rod-type) is interconnected and large enough to host the quasi-graphitic nanodomains derived from the filling PTCDA. As a result, nanodomains with layered graphitic structure instead of nanocavity was introduced into the CMK8GC and then no plateau occurs.
As shown in Fig. S24, CMK8GC electrode derived from the mesopore-dominated carbon host (CMK-8) displays no plateau capacity during discharge-charge process. Figure S25 shows that the ACGC1050 electrode can reach about 118 mAh g -1 even at 2000 mA g -1 . Moreover, about 97.3% of the initial capacity can be maintained after 200 cycles at 50 mA g -1 .
As displayed in Figs. S26-S28, if the cell voltage is linearly proportional to τ 1/2 , the diffusion coefficient can be calculated from the GITT potential profiles by Fick's second law with the following equation: The density of carbon was calculated according to the following equation: where ρ (g cm -3 ) is the density of carbon, Vtotal (cm 3 g -1 ) is the total pore volume measured from the N2 isotherm, ρcarbon is the true density of carbon (2.08 g cm -3 ).
For the GITT tests, the cell was discharged/charged at C/10 with a current pulse duration of 0.5 h and an interval of 1 h.S1,S2 As shown in Fig. S29, the D-band (1340 cm -1 ) and G-band (1580 cm -1 ) keep the same peak location and no new peaks appear during the sodiation/desodiation process, suggesting no intercalation of Na + into/from graphitic interlayers. Figure S30 shows that no resonance peak can be observed at around 750 ppm, indicating no quasi-metallic Na metal existing in the discharged electrode. Figure S31 schematically illustrates the aforementioned "adsorption/pore-filling" mechanism in a sodiation process, i.e. the sloping region above 0.1 V results from the adsorption of Na + on the surface sites, while the plateau region below 0.1 V originates from the pore-filling process of Na + into the ultra-micropores.
As demonstrated in Fig. S32, compared with the high reversible capacity at a small current density (0.1 mA cm -2 ), ~ 53.1 % of capacity (3.26 mAh cm -2 ) can still be maintained at a much higher current density (0.5 mA cm -2 ), indicating the superior rate capability of the fabricated thick electrode. Figure S33 indicates that similar diffusion kinetics of Na + inside the thick electrode at various temperatures can be observed from the potential-dependant DNa + profiles, suggesting a satisfying low-temperature performance.
As shown in Table S1, the specific surface area of ACGCx calculated from N2 adsorption/desorption test decreases with the increasing carbonization temperature. SBET of ACGCx decreases with increasing temperature. Table S2 shows the exploration about the effect of the carbonization on the structure of AC. As shown in Fig. S3 and Table S2, the pore structure of AC maintains well after carbonization under 1050 ℃. Table S3 show the skeletal (true) density data of the ACGCx materials. Specifically, the skeletal (true) density data were recorded on an AccuPyc II 1340 analyzer using Helium as analysis gas. The skeletal (true) density monotonically increases from 1.89 g cm -3 (ACGC750) to 2.14 g cm -3 (ACGC1200) as the temperature elevated. Table S4 indicates that the R value of ACGCx increases with the carbonization temperature, where a lower R value suggests a lower degree of graphitization or more defect sites.
As shown in Table S5, the initial coulombic efficiency (ICE) value of various samples ranges from 68.1 to 87.9%. Table S6, after the molten diffusion-carbonization process, the obtained ACGC, LCGC, HCGC and CMK8GC and HCGC materials suffer from a huge reduction in specific surface area as compared with AC, LC, HC, and CMK8, respectively.

As indicated in
As shown in Table S7, LC (low SBET carbon, 224 m 2 g -1 ) and HC (high SBET carbon, 2939 m 2 g -1 ) were also selected as carbon host for comparison. After the molten diffusioncarbonization process, the obtained LCGC and HCGC materials suffer from a huge reduction in specific surface area, which is consistent with ACGC.
As shown in Table S8, a high reversible capacity of 125 mAh g -1 can still be delivered at 2000 mA g -1 , which reveals a superior performance compared to that of the previously reported hard carbon anode materials.S3-S13   Notes: ID/IG ratio is calculated from the ratio between the peak area of the D band and G band in the Raman spectra. A smaller ID/IG ratio indicates a higher degree of graphitization.

Fig. S18
Relationship between the sloping capacity contribution and the ID/IG ratio calculated from Raman spectra of ACGCx

Fig. S19
Relationship between the ICE value and the BET specific surface area obtained from N2 adsorption/desorption test   If the cell voltage is linearly proportional to τ 1/2 , the diffusion coefficient can be calculated from the GITT potential profiles by Fick's second law with the following equation: The density of carbon was calculated according to the following equation: where ρ (g cm -3 ) is the density of carbon, Vtotal (cm 3 g -1 ) is the total pore volume measured from the N2 isotherm, ρcarbon is the true density of carbon (2.08 g cm -3 ).
For the GITT tests, the cell was discharged/charged at C/10 with a current pulse duration of 0.5 h and an interval of 1 h [S1, S2].     [a] Micropore specific surface area determined by the t-method by N2 adsorption branch at 77 K.
[b] External specific surface area determined by the t-method external surface area [c] Micropore volume detemined by the HK method by N2 adsorption branch at 77 K [d] External volume determined by the summation of mesopore volume from the DFT method and macropore volume calculated from adsorption curve by BJH method [e] Total volume determined by the summation of the micropore, the external volume [a] Micropore specific surface area determined by the t-method by N2 adsorption branch at 77 K.
[b] External specific surface area determined by the t-method external surface area [c] Micropore volume detemined by the HK method by N2 adsorption branch at 77 K [d] External volume determined by the summation of mesopore volume from the DFT method and macropore volume calculated from adsorption curve by BJH method [e] Total volume determined by the summation of the micropore, the external volume   [a] Micropore specific surface area determined by the t-method by N2 adsorption branch at 77 K.
[b] External specific surface area determined by the t-method external surface area [c] Micropore volume detemined by the HK method by N2 adsorption branch at 77 K [d] External volume determined by the summation of mesopore volume from the DFT method and macropore volume calculated from adsorption curve by BJH method [e] Total volume determined by the summation of the micropore, the external volume