MXenes have gained tremendous attention in supercapacitor^[1], lithium battery^[2], electromagnetic interference shields^[3] and other energy-related fields. They were firstly reported by Michael Naguib via selectively etching from MAX phases^[4] (where M is a transitional metal, A is a III or IV A-group element and X is carbon or nitrogen). The resulting 2D materials, i.e., MXenes, have many chemical and structural forms with high intrinsic electronic/ionic conductivities. When applied to supercapacitors, the fast surface redox makes them possess large energy. However, although much effort have been made to obtain superior gravimetric and volumetric capacity of MXenes^{[1, 5]}, the mass loading of MXenes still needs to be further improved in order to get a high areal capacity and meet practial applications in large scale.

In 2017, Chen, et al^[6] reported a fascinating work to utilize natural basswood structure for an all-wood-structured asymmetric supercapacitors design. The electrodes were derived from natural wood and inherited its unique anisotropic structure which has numerous open channels along the growth direction. The wood based devices displayed high desirable thickness (up to ~1 mm) with high mass loading of 30 mg/cm². With these unique features, such as low tortuosity, high electronic conductivity, enabling high mass loading, low cost and environmentally friendly, the wood-structured electrode represents a promising way to realize high power density and high energy density for green energy storage device^[7,8,9].

Herein, we reported a facile way to make a high mass loading NaOH-Ti₃C₂T_x/wood carbon electrode for supercapacitors. The resulted electrode exhibited a high areal capacity of 1983 mF/cm² at 2 mV/s with a high Ti₃C₂T_x mass loading of 17.9 mg/cm², which surpass three times than that of NaOH-Ti₃C₂T_x reported before^[10].

Ti₃C₂T_x (T=O, OH) was synthesized according to the paper reported before^[10]. About 80 mg Ti₃AlC₂ powder (400 mesh (38 μm), purchased from 11 Technology Co. Ltd.) was added into the 35 mL NaOH solution (27.5 mol/L, the water was heated to boiling to reduce the concentration of oxygen) contained in a 50 mL autoclave in argon (Ar) atmosphere. Then the autoclave was heated at 270 ℃ in an electric thermostatic drying oven for 12 h. Finally, the resultant suspension was filtered with deionized water washing for several times. The obtained powder was dried at 60 ℃ for 12 h.

Natural basswood blocks were cut perpendicularly in the growth direction to afford wood slices. Then the slices were pre-carbonized in air at 250 ℃ for 6 h, followed by carbonization in an argon flow at 1000 ℃ for 6 h. The carbonized slices were carefully polished with 2000 grit sandpaper to obtain a thickness of 700 μm and residual carbon was removed with ultrasonic washing by deionized water and ethanol for three times. To further reduce the weight and activate the wood carbon, the resultant wood carbon slices were activated in a carbon dioxide (CO₂) flow at 750 ℃ for 1 h in a gas flow of 0.2 L/min.

Ti₃C₂T_x was loaded into the channels via vacuum- assisted infiltration. Briefly, 160 mg amount of prepared Ti₃C₂T_x powder was dispersed in 10 mL ethanol via sonication for 30 min. The as-prepared wood carbon was placed on a filter funnel and then the obtained homogeneous ink was dripped into the wood carbon matrix along the microchannels after vacuum filtration using a turbo pump. The wood carbon filled with Ti₃C₂T_x was treated by freeze-drying for 48 h. The mass ratio of Ti₃C₂T_x with wood carbon was ~ 1 : 2.4.

XRD test was conducted using a Rigaku Ultima IV Powder Diffractometer (Cu Kα radiation, sweeping speed: 5(°)/min). TEM analysis was performed on a JEM-2100F transmission electron microscope (200 kV, JEOL, Japan). SEM studies were carried out on a Mira3 scanning electron microscope (5 kV for morphology observation, Tescan, Czech).

All electrochemical measurements were performed using a traditional three electrode system with 1 mol/L H₂SO₄ solution as electrolyte. Ti₃C₂T_x/wood carbon electrode, carbon rod, and Hg/Hg₂SO₄ in saturated K₂SO₄ aqueous solution were used as working electrode, counter electrode, and reference electrode, respectively. Cyclic voltammograms (CVs) and galvanostatic charge- discharge (GCD) analysis were performed on an electrochemical workstation (Biologic VMP3). The scan rates for CV analysis were 1, 2, 5, 10 and 20 mV/s. The scan range (-0.5 ~ 0 V) was determined by series of CV scans at 10 mV/s.

Ti₃C₂T_x was synthesized from bulk material Ti₃AlC₂via hydrothermal method (27.5 mol NaOH, Ar atmosphere, 270 ℃). The raw material Ti₃AlC₂ was shown in Fig. 1(a). The bulk Ti₃AlC₂ material was made from Ti₂AlC and TiC by a hot pressing sintering method^[11].

There were some impurities such as Al₂O₃ (See XRD patterns of Ti₃AlC₂, Fig. 1(c)) during the synthesis process. After alkali treatment, sheet structure emerged in obtained Ti₃C₂T_x (Fig. 1(b)), which indicated the successful etching of Al atoms. It was also supported by X-ray diffraction (XRD) that the rising peak at 2θ≈7.5° corresponding to (002) plane of Ti₃C₂T_x. The morphology of NaOH-Ti₃C₂T_x was different from accordion-like HF-Ti₃C₂T_x via high concentration (50wt%) HF treatment but similar to low concentration (5wt%) HF treatment^[11]. It is noted that the position of (002) peak in NaOH route is similar to HF etching route followed by sodium intercalation^[12], which indicates that NaOH treatment has the same effect. Furthermore, we also performed Transmission electron microscopy (TEM) analysis to observe the morphology of Ti₃C₂T_x. Lamellar stripes can be observed in the edge of Ti₃C₂T_x sheets and the interlayer space was ~1.2 nm. The space was larger than that of hydrofluoric acid (HF) method, which can be explained by sodium intercalation during the hydrothermal process.

SEM was used to observe the structure of wood carbon. Fig. 2(a) showed the top-view of wood carbon, where multi big channels (30-60 μm) were surrounded by numerous small channels (5-10 μm). From the cross- sectional view (Fig. 2(b)), we can clearly observe that almost all channels were aligned straight from top to bottom. Furthermore, there were some mesopores (high magnification image in Fig. 2(b)) existed in channels, which were beneficial to diffusion of electrolyte. Next, Ti₃C₂T_x ink was made via sonication in ethanol for 30 min and then dripped into the wood carbon with a vacuum filtration method. SEM images showed that multilayer Ti₃C₂T_x was successfully injected (Fig. 2(c, d)). However, Ti₃C₂T_x was not uniformly distributed in the channels of wood carbon. Some channels were filled with Ti₃C₂T_x, while some were empty, which may not take full use of wood structure and every single Ti₃C₂T_x.

Since wood displayed anisotropic structure with vertical micro-channels serving as the high transport pathway of ion transport in the electrode, we thus tested the supercapacitor performance of Ti₃C₂T_x/wood carbon. The supercapacitor performance was evaluated using a three-electrode cell with 1 mol/L H₂SO₄ aqueous solution serving as the electrolyte. To the best of our knowledge, Ti₃C₂T_x is stable from -1.0-0 V vs. Hg/HgSO₄ while the wood carbon is stable from -0.5-0.5 V vs. Hg/HgSO₄ according to our experiments (Fig. 3(a)). Therefore, -0.5-0 V was chosen as this potential range which was suitable for both of them. Fig. 3(b) showed the CV curves of Ti₃C₂T_x/wood carbon. Different scanning rates ranging from 1 to 20 mV/s were applied. For wood carbon, the CV curves were nearly like rectangle shape which indicated fast charge transfer in the electrode and the wood carbon exhibited high areal capacity of 1030 mF/cm² at 2 mV/s. For Ti₃C₂T_x/wood carbon (mass loading: 17.9 mg/cm²), the area of CV curves were relatively larger than that of bare wood at the same potential range and the performance was 1983 mF/cm² at 2 mV/s, representing good values among Ti₃C₂T_x electrodes reported before^[10]. Even at 20 mV/s, the capacity of Ti₃C₂T_x/wood carbon electrode still maintained 1539 mF/cm². GCD curves of Ti₃C₂T_x/wood carbon were showed in Fig. 3(c). The linear GCD curves with triangular shape suggested ideal capacitive behavior, while the repeated GCD curves at 10 mA/cm² for 10⁴ cycles indicated good stability (maintained ~92%) for Ti₃C₂T_x/wood carbon electrode (Fig. 3(d)).

In conclusion, we have demonstrated wood-based Ti₃C₂T_x electrode a unique structure for supercapacitor to achieve high areal capacity. Open channels of wood benefit fast and efficient transportation of ions. When Ti₃C₂T_x was combined with wood carbon, a high areal capacity (1983 mF/cm² at 2 mV/s) of with high mass loading (17.9 mg/cm²) was achieved. This work also inspires us to explore other natural structure, which may possess fascinating properties in combination with other materials.