4D printing of MXene hydrogels for high-efficiency pseudocapacitive energy storage

2D material hydrogels have recently sparked tremendous interest owing to their potential in diverse applications. However, research on the emerging 2D MXene hydrogels is still in its infancy. Herein, we show a universal 4D printing technology for manufacturing MXene hydrogels with customizable geometries, which suits a family of MXenes such as Nb2CTx, Ti3C2Tx, and Mo2Ti2C3Tx. The obtained MXene hydrogels offer 3D porous architectures, large specific surface areas, high electrical conductivities, and satisfying mechanical properties. Consequently, ultrahigh capacitance (3.32 F cm−2 (10 mV s−1) and 233 F g−1 (10 V s−1)) and mass loading/thickness-independent rate capabilities are achieved. The further 4D-printed Ti3C2Tx hydrogel micro-supercapacitors showcase great low-temperature tolerance (down to –20 °C) and deliver high energy and power densities up to 93 μWh cm−2 and 7 mW cm−2, respectively, surpassing most state-of-the-art devices. This work brings new insights into MXene hydrogel manufacturing and expands the range of their potential applications.


Reviewer #1 (Remarks to the Author):
The authors report a 4D printing approach for manufacturing MXene hydrogels. This strategy shows remarkable universality, which suits not only the most commonly studied Ti3C2Tx MXene but also Nb2CTx and Mo2Ti2C3Tx. These three MXenes with different atomical structures and properties can represent a family of MXenes. It has never been achieved in previous reports. Besides, the printed MXene hydrogel architectures are precise and customizable, which is unlikely to realize in traditional model-dependent methods. The formation mechanism of MXene hydrogel is also clearly investigated. In capacitive application, the thickness/mass loading-independent rate capability of MXene hydrogel is very impressive, which surpasses the most advanced electrodes ever reported. The areal capacitance of MXene hydrogel also meets the commercial requirements even at a high scan rate of 1 V s-1. The further printed micro-supercapacitor delivers outstanding capacitive performance and prominent low-temperature tolerance. This work is very interesting, which will advance the development of both MXenes and hydrogels and attract a wide range of attention in different fields, including additive manufacturing, material science, bioengineering, electrochemistry, etc. It is recommended to be published on Nature Communications after minor revision. Followings are some comments: 1. The electrical conductivity of Ti3C2Tx hydrogel is much higher than the other two MXene hydrogels, the authors should discuss the possible reasons, and present the electrical conductivity values of three MXene films for comparison. In addition, electrical conductivity is crucial in various applications, such as biomedical, sensing, and electrochemistry. The comparison of the electrical conductivity of the MXene hydrogels with other reports should be presented, as well as their application fields. This will reveal the other potential applications of MXene hydrogels. 2. How about the capacitive performance of pure PETOT:PSS hydrogel and Ti3C2Tx MXene film? 3. In the Introduction section (Line 58), the authors claim that the 4D printing technology is scalable, but only several architectures are shown in this manuscript. The authors should manufacture more hydrogel units to demonstrate this property. 4. Five kinds of architectures are shown in Figure 2e, but only 2 corresponding videos are presented in Supplementary Movies. It would be better to show all the relating videos. 5. The micro-supercapacitors exhibit excellent low-temperature tolerance and the PVE-EG-H2SO4 electrolyte is a key contributor. How about the ionic conductivity of this electrolyte in comparison with other similar electrolytes? This will advise the future development of lowtemperature devices. 6. Some minor errors are found, such as "10 mins" (Line 399) and "20 min" (Line 405 (e.g. Adv. Energy Mater.2020, 10, 1903794;Energy Environ. Sci., 2019, 12, 96--115). 3. In the "probing self-assembly mechanism" section, the authors mentioned that "suggesting the conformation change from benzene structure to quinoid structure and thus elongated conjugation lengths of PEDOT+ chains." The authors are suggested to add necessary citation to rationalize the discussion. 4. The electrical conductivity of Ti3C2Tx HGs can reach 1548 S m-1, which is far more than the cases of Nb2CTx and Mo2Ti2C3Tx HGs. The authors should give detailed reasons in the manuscript. 5. More experimental details could be included in the manuscript. For instance, the electrolyte amount used in the devices should be provided, which is an important parameter to evaluate the electrochemical performance of supercapacitors. And the thickness of electrodes should be provided in the Methods section. 6. In Figure 2b, the green line cannot be found, the authors should clearly check the figures. 7. In Figure 3h, the scale range of x-axis and y-axis is inconsistent. In addition, have the EIS data been normalized by mass or surface are of electrode? 8. In page 13 of main text and Figure 4d, the author mentioned the energy density and power densities. The reviewer wonders that are these values calculated based on total mass of cells or only of electrode.
We thankfully acknowledge all the reviewers for their time and valuable comments on our manuscript. The revisions have been highlighted in yellow in the revised manuscript. Please find below our point-by-point responses.

Reviewer #1:
The authors report a 4D printing approach for manufacturing MXene hydrogels. This strategy shows remarkable universality, which suits not only the most commonly studied Ti3C2Tx MXene but also Nb2CTx Figure R1 (Supplementary Figure 13)). The electrical conductivities of Nb2CTx film, Ti3C2Tx film, and Mo2Ti2C3Tx film are 382 S m −1 , 58,149 S m −1 , and 3,018 S m −1 , respectively.
These values are an order of magnitude higher than their corresponding hydrogels, suggesting that the electrical conductivity of MXene hydrogels is highly dependent on the electrical conductivity of MXenes. The Ti3C2Tx film has the highest conductivity and endows the Ti3C2Tx hydrogel with the highest conductivity. Conversely, Nb2CTx hydrogel shows the lowest conductivity. In the revised manuscript, we have added a statement to explain the electrical conductivity difference between the three MXene hydrogels: "The conductivity difference between these hydrogels should be ascribed to the intrinsic conductivity difference of the three MXenes ( Supplementary Fig. 13)." Following the reviewer's suggestion, we have created a new table (Supplementary Table 2) to list the electrical conductivity and application fields of various hydrogel materials. The electrical conductivity values of our 4D-printed MXene hydrogels are higher than most hydrogels, revealing the potential of our MXene hydrogels for wide applications, such as sensors, bioelectronics, electromagnetic interference shielding, and electrochemical energy storage. (   Response: The authors would like to thank the reviewer for these valuable and insightful comments. Following the comments, we have added more discussion and data to address all the raised suggestions. (1) For the novelty concerns, our work shows great advances over previous reports and realizes some achievements that have never or hardly been achieved before: Thirdly, the geometry of the previously reported MXene hydrogels strongly depends on the shape and size of molds, which is unlikely to meet the requirements of complexity and precision in many scenarios, especially in the context of the rapid development of portable electronics. In this work, our 4D printing technology allows the precise and mold-free fabrication of various complex MXene hydrogels, such as microlattice, rectangular hollow prism, Chinese knot, "CRANN" logo, and micro-supercapacitor (MSC) units. And series of substrates suit, e.g., glass slide, cloth, and PET film ( Figure R4 (Figure 2e)). This 4D printing technology significantly surpasses the mold-dependent methods and shows great potential for modern electronics. Fourthly, benefiting from the large specific surface area, high electrical conductivity, and hydrophilic property, ultrahigh capacitance (3.32 F cm −2 at 10 mV s −1 and 233 F g −1 at 10 V s −1 ) and unprecedented mass loading/thickness-independent rate capabilities are achieved for our Ti3C2Tx hydrogel electrode, which surpasses most state-of-the-art electrode materials (Supplementary Table 5). Particularly, in commercial applications, the typical electrode areal capacitance is 0.6 F cm −2 (horizontal dash line in Figure R5 (Figure 3g)), and to date, only the liquid-crystalline Ti3C2Tx (Nature 557, 409-412 (2018)) and 1T-MoS2 (Nat. Nanotech. 17, 153-158 (2022)) electrodes have achieved this value at 1 V s −1 . Our MXene hydrogel electrodes not only meet the commercial requirements, but also possess X2 higher areal capacitance than liquidcrystalline Ti3C2Tx and 1T-MoS2 electrodes at both 1 and 2 V s −1 under similar mass loadings (6.16−6.6 mg cm −2 ) ( Figure R5 (Figure 3g)). This is a very large progress and demonstrates the potential of our MXene hydrogels for practical applications. The further fabricated MSCs also 12 deliver larger areal capacitance and higher energy/power densities than most printed MSCs, including the 3D-printed MXene MSCs. The low-temperature performance of our MSC is excellent as well ( Figure R6 (Figure 4c,d,f)).  To conclude, our work has addressed many difficulties that have not been solved and made great achievements over previous reports in both MXene hydrogels manufacturing and electrochemical energy storage. We think our work is novel and has met the high level of Nature
(2) To compare the 4D printing with 3D printing and highlight the advantages of 4D printing, we have added a new paragraph in the Introduction section: "Additive manufacturing, or 3D printing, offers an efficient approach to realizing the precise, mold-free, and low-cost fabrication 14 of complex objects by layer-by-layer deposition of material 21 . With the introduction of the fourth dimension of time, 4D printing (3D printing + time) emerged 22 . It not only inherits all merits of 3D printing but also allows the static objects created by 3D printing to change their shape, property, or functionality over time when exposed to specific external stimuli (e.g., heat, light, water, pH) 23 , endowing the printed objects with new features. However, no related works on MXene hydrogels were ever reported." (Lines 57-63, Pages 2-3)

Compared to 3D printing technology, what are the advantages of MXene HGs by 4D printing?
The authors should give more discussion in the introduction.

Response:
The 3D-printed MXene patterns or architectures are usually in sol states, which are essentially inks or condensed dispersions but with customizable geometries (e.g., Adv. Mater. 31, 1902725 (2019); ACS Nano 14, 640-650 (2019)). Once immersed in water, they will be easily redispersed. Therefore, freeze-drying is indispensable to maintain the shape of 3D-printed objects before application. Besides, the final products are aerogels, not hydrogels. In contrast, our 4D-  Figure R7 (Supplementary Figure 9). Photographs of 4D-printed Ti3C2Tx hydrogel microlattice and rectangular hollow prism before and after shaking for ~14 s. After shaking, the two hydrogels retained their integrity. Figure R8 (Supplementary Figure 10). Photographs of 3D-printed Ti3C2Tx sol microlattice before and after shaking for ~8 s. After shaking, the sol microlattice broke into fragments. It is worth noting that, there are already some electrostatic attractions between the negatively charged MXenes and positively charged PEDOT + chains and protons, which protect this sol microlattice from complete redispersion in water. The pure Ti3C2Tx sol architectures that only possess weak van der Waals interactions 4, 5 will be completely redispersed after shaking.
Response: We agree with the reviewer's point. To make this expression more precise, we have replaced the word "ultrahigh" with "high". (Line 25, Page 1) We read these two papers carefully and there are two main reasons that lead to higher power density than ours. Firstly, our device has a thickness of 4 mm; the devices in the mentioned papers are, in contrary, very thin (almost all of them are less than 0.5 mm). Secondly, our device is currentcollector free; metal current collectors or high-conductive metal fillers (e.g., Ag nanowire and Au) were employed in the cited papers.
3. In the "probing self-assembly mechanism" section, the authors mentioned that "suggesting the conformation change from benzene structure to quinoid structure and thus elongated conjugation lengths of PEDOT+ chains." The authors are suggested to add necessary citation to rationalize the discussion. Response: The observation is highly appreciated and to amend this issue, we measured the electrical conductivity of three MXene films in the revised version and found the electrical conductivity of MXene hydrogels is highly dependent on the electrical conductivity of MXenes.

Response:
To avoid duplication, please see our response to Reviewer #1's 1 st comment and Figure R1 (Supplementary Figure 13)  (2) The photograph, schematic, and model number (Line 500, Page 19) of the Swagelok cell that used for three-electrode tests are provided ( Figure R9 (Supplementary Figure 15)). (3) The schematic of designing 4D-printed Ti3C2Tx hydrogel MSC is provided ( Figure R10 (Supplementary Figure 21)). Figure 21). Schematic of designing 4D-printed Ti3C2Tx hydrogel MSC. The volume percentage of electrodes to the whole MSC is ~70 vol.%, and the gap between electrodes is ~30 vol.%. Because the electrodes are highly porous, they can absorb almost as much electrolyte as their volume. Thus, the volume of the gel electrolyte added should be at least the same as the volume of MSC (including both the electrodes and the gap). To ensure the complete infiltration of electrodes and maximize the electrochemical performance of MSC, the total volume of the cast PVA-EG-H2SO4 gel electrolyte was set to 120 vol.% of the MSC.