MXene-based Zn-ion hybrid supercapacitors: Effects of anion carriers and MXene surface coatings on the capacities and life span

,


Introduction
The high demand on safe, clean, low-cost, and renewable energy sources ensures a high priority for research on rechargeable energy storage device.Supercapacitors are fast rechargeable devices, however, the low energy density, especially for electric double layer supercapacitors, limits their use [1][2][3][4][5].On the other hand, lithium-ion batteries possess a higher energy density, but a slower rate performance, as compared to supercapacitors.Moreover, lithium has a low abundance and there are safety concerns on the organic electrolyte flammability [6,7].Accordingly, other systems based on monovalent or divalent cations have been investigated, such as Na + , K + , Mg 2+ , Ca 2+ and Zn 2+ [8][9][10].Among them, the Zn-ion charge storage devices have attracted a great attention [11][12][13] due to 1) suitable standard electrode potential (− 0.76 V vs. standard hydrogen electrode), 2) high theoretical capacity (825 mAh g − 1 ), and (3) zinc metal being abundant, nontoxic, and cheap material (2 USD per kg) [14,15].
Zn-ion hybrid supercapacitors (ZHSC) belong to the group of promising energy storage devices due to their high energy density and power density compared to conventional symmetric or asymmetric supercapacitors [16,17].Carbon based electrodes such as activated carbon [18,19], porous carbon [20][21][22][23], and hollow carbon spheres [24] are the most studied positive electrode materials for ZHSC.However, these carbon-based electrodes were prepared using a conventional cast method which involves the use of carbon black and organic binder additives, which leads to a decrease in the specific device capacitance.In addition, the energy storage mechanism for carbon materials is mainly limited by the electric double layer capacitance (EDLC) [25].Alternatively, transition metal carbides known as MXenes [26], are a class of 2D materials which have been tested for energy storage devices including ZHSC [27][28][29][30].The general formula of MXenes is M n+1 X n T z , where M = transition metal (e.g.Ti, Mo, V, Nb, Ta, etc.), X = C and/or N, and T = surface termination groups (e.g.-OH, -F, -Cl, or -O).MXenes are synthesized from their parent 3D MAX phase with the general formula M n+1 AX n , via selective chemical etching of (A) element where A typically is Al.MXenes possess a unique combination of properties, such as high conductivity, tunable mechanical properties, and high specific capacitance [31,32].In particular, titanium based MXenes, Ti 2 CT z and Ti 3 C 2 T z [29,30], were used as positive materials for ZHSC.A MXenebased ZHSC device featured a capacitance of about 132 F g − 1 at 0.5 A g − 1 , in addition the devices after cycling can be recycled through complete degradation in phosphate buffer saline within about 7.25 days [27].Notably, the energy density of the previously reported MXenebased ZHSC devices was limited by its low capacitance, and therefore modifications are required to improve the performance.
Various electrode manufacturing processes have been developed to increase the accessible capacitance/capacity of the MXene electrodes.For example, cetyltrimethylammonium bromide (CTAB)-pillared Ti 3 C 2 T z produced 53 mAh g − 1 at 0.2 A g − 1 [30], Ti 3 C 2 T z composites with reduced graphene oxide delivered 50 mAh g − 1 at 0.4 A g − 1 [25], and Sn 4+ -pre-intercalated-Ti 2 CT z on porous carbon spheres offered 114 mAh g − 1 at 0.5 A g − 1 [29].Furthermore, halogenated Ti 3 C 2 T z with Br − and I − terminations showed a capacity of 135 mAh g − 1 at 0.5 A g − 1 [33].In addition, modified Nb 2 CT z MXene featured a capacity of about 150 mAh g − 1 [34] and was employed as host for confinement of I 2 in I 2 -Zn batteries [35].More recently, composite MXene electrodes, including Mo 1.33 CT z and Ti 3 C 2 T z , were used as a positive electrode for a Zn hybrid supercapacitor in 3 M Zn (CF 3 SO 3 ) 2 solution, delivering a promising specific capacity up to 159 mAh g − 1 [36].The improved performance was explained by a high electronic conductivity and an increased number of the accessible active sites owing to a less compact stacking and curved layer morphology of the mixed MXene sheets [36].However, the Zn(CF 3 SO 3 ) 2 salt is quite expensive, and the exploration of other cheaper Zn salts, such as ZnSO 4 and ZnCl 2 , is thus highly motivated.
The use of a Zn metal anode has several problems, such as uncontrolled dendrite formation, which can result in a short circuit of the Zn cells after just a few cycles, especially at high current densities [13,37,38].In addition, the use of Zn metal is hindered by parasitic hydrogen evolution reaction (HER) during long-term cycling [13,37,38].The HER causes a decrease in the coulombic efficiency and raises the medium pH [39], which leads to formation of passivating side products such as ZnO and Zn(OH) 2 .The former reduces the conductivity and increases the polarizability, which deteriorate the long-term stability [13,37,38].Accordingly, various approaches have been employed to solve these problems.Coating the Zn surface with a surface layer can provide a good solution [13,37,38], using e.g.graphite [40], Cu nanowires [41], or Ti 3 C 2 T z MXene coatings [42]; alternatively, Ti 3 C 2 T z @Zn paper electrode can also address the dendrite growth problem [43].However, these approaches were mainly studied in ZnSO 4 electrolyte, and the anion carriers of aqueous Zn electrolytes can affect the efficiency of Zn plating-stripping as well as the accessible capacity of the MXene// Zn full cells [13,44,45].For example, the MXene electrodes suffers from spontaneous formation of Zn(OH) 2 /ZnSO 4 flakes on the electrode surface immediately after assembling cells using ZnSO 4 solution [46].Accordingly, exploration of the effect of coating the Zn electrodes with Ti 3 C 2 T z MXene for use in aqueous electrolytes rather than ZnSO 4 is highly motivated.Zinc chloride is a cheap Zn salt with high solubility in water; however, the Zn cells with a low concentration ZnCl 2 solution suffer from fast passivation of the Zn electrode surface due to the formation oxide and/or hydroxide deposits, causing a poor platingstripping [11,47].The use of highly concentrated ZnCl 2 solution can, however, address the problems associated with the use of low concentration solutions [48].
In this paper, we report on the effect of anion carriers (Cl − , I − , SO

Synthesis, morphology, and structure of the composite MXene films
The Mo 1.33 CT z and Ti 3 C 2 T z MXene suspensions were prepared through chemical etching of their parent i-MAX and MAX phases, (Mo 2/ 3 Sc 1/3 ) 2 AlC and Ti 3 AlC 2 , respectively, the details are shown in the experimental techniques section [49,50].The composite MXene films were prepared by mixing Mo 1.33 CT z and Ti 3 C 2 T z MXene suspensions in a weight ratio of 3:1, as this ratio has been shown to be optimal for obtaining a uniform flexible electrode with high capacity [36,51].After mixing, the suspension was hand shaken for 30 s, and a freestanding film was obtained by vacuum filtration.For further details, see Ref. [51].Fig. 1a shows a schematic illustration of the synthesis process.The as prepared freestanding films were flexible (see inset in Fig. 1b) and assigned as a 3Mo:1Ti film.The SEM cross-section, see Fig. 1b-d, showed that the mixed MXene films possess a typical layered structure with curved layer morphology and less compact stacking of the MXene sheets, in agreement with previous reports on mixed MXenes [51,52].The individual MXene films featured a more compact staking of the MXene sheets.(see Fig. S1a and b).The X-ray diffraction (XRD) pattern of the as-prepared MXene film (see Fig. S1c) displayed a low-angle peak at around 5.7 • , corresponding to a d-spacing of about 15.5 Å, in agreement with previous reports [36,51].This d-spacing value is larger than those of the low-angle peaks of the Ti 3 C 2 T z MXene (~11.7 Å), and the Mo 1.33 CT z (~15.0Å) [51], suggesting that the mixed MXene film offer more accessible sites for ions insertion/intercalation during the electrochemical test.

Effect of anion carrier on the electrochemical performance of mixed MXene//Zn cells
The effect of anion carrier on the electrochemical performance of the MXene//Zn cells was investigated using different zinc solutions with concentration of about 3 M.The initial screening tests involved aqueous solutions of the salts ZnCl 2 , ZnI 2 , ZnSO 4 , Zn(CF 3 SO 3 ) 2 (abbreviated as Zn-Triflate), and Zn di[bis (trifluoro methyl sulfonyl) imide] (abbreviated as Zn(TFSI) 2 ).Cyclic voltammetry and constant current measurements were used to examine the electrochemical performance in different electrolytes (see Fig. 2 and Table 1).The potential window was selected to be 0.01-1.3V vs. Zn 2+ /Zn [36], with the exception of the ZnI 2 solution where the upper cut-off potential window was limited to 1.1 V due to the redox activity from the I 2 moiety occurring around 1.2 V (see Fig. S2a).
The operation of MXene//Zn cells during charging and discharging modes is schematically presented in Fig. 2a and b, respectively.In the charging mode, the Zn 2+ is plated (reduced) at the surface of the Zn foil, whereas the anions migrate and physically adsorb on the surface of the mixed MXene electrode.In the discharging mode, the Zn metal is stripped (oxidized) from the Zn anode surface producing Zn 2+ in the electrolyte, while the anions migrate from the mixed MXene electrode to the electrolyte.The Zn plating-stripping was monitored during the MXene//Zn cells operation using an Ag/AgCl electrode (see Fig. 2c).The overpotentials in ZnCl 2 and ZnI 2 were quite low, 59 and 12 mV, respectively, as compared to the overpotentials in Zn(CF 3 SO 3 ) 2 (130 mV), Zn(TFSI) 2 (155 mV), and ZnSO 4 (180 mV).The shift in the platingstripping potential observed in ZnSO 4 solution can be attributed to the formation of Zn(OH) 2 /ZnSO 4 flakes on the surface of the Zn electrode immediately after assembling cells with ZnSO 4 solution [46].The CVs of MXene//Zn cells featured a typical pseudocapacitive shape in different electrolytes at scan rate of 0.5 mV s − 1 , with a pair of redox peaks located at a potential of 0.30-0.35V for reduction and 0.45-0.50V for oxidation (see Fig. 2d).The observed redox reaction should also involve insertion/ deinsertion of Zn 2+ ions and/or protons between the MXene flakes in order to maintain electric neutrality [28,53,54].Also, it was shown that the carbon-based electrodes can possibly adsorb anion as well as Zn 2+ cations at high (>0.75V) and low (<0.75 V) cell voltage regions respectively [55].However, the adsorption on the MXene electrode surface affected by the surface termination groups acquiring the surface negative charge [56][57][58][59].When the scan rate was raised to 10 mV s − 1 , the peak-to-peak separations (ΔU p ) in the case of SO 4 2− , CF 3 SO 3 − , and TFSI − anions were larger than those in the presence of halide ions Cl − and I − (see Fig. 2e).The latter observation could be attributed to the different mobilities of the anions.At a higher scan rate of about 100 mV s − 1 the CV shape became highly distorted, which can be attributed to a combination of the increased iR drop and diffusion limitations (see Fig. S2b).The decline observed in 3 M ZnCl 2 can be explained by passivation of the Zn electrode surface as a result of the formation of side products (oxide and/or hydroxide side deposits) causing a poor platingstripping [48].It should be noted that all MXene//Zn cells need to be precycled for 25-50 cycles at scan rate 10 mV s − 1 prior to cycling at different scan rates (see Fig. S3) [36].
It worth noting that the ionic conductivity of the electrolytes should be considered when comparing the performance of the different electrolytes (Zn(CF 3 SO 3 ) 2 -3.47 S cm − 1 [60,61], Zn(TFSI) 2 [61], ZnSO 4 [61], ZnCl 2 ~ 10 − 1 S cm − 1 [48,61], and ZnI 2 [62]).The electrochemical impedance spectroscopy (EIS) measurements were also conducted to follow the changes in the resistance of MXene//Zn cells before and after precycling in different anion carriers.It can be seen immediately that the halide anion carrier showed the lowest charge transfer resistance (R ct ), which is determined from the radius of the semicircle in Nyquist plots (see Fig. 2f and i).In addition, the charge transfer resistance in different anion carriers decreased significantly after the precycling, suggesting that the structure of MXene films became more open after precycling.Also, the internal resistance for MXene//Zn cells, which is determined from the intersection with the x-axis in the Nyquist plot, decreased after precycling, which can be explained by the removal of any oxide deposits remaining on the surface of the Zn electrodes.
The galvanostatic charge-discharge curves showed symmetric profiles with a sloping plateau, indicating the pseudocapacitive behavior of the charge storage mechanism.As the current density increased, the iR drop became more pronounced, which is in line with the magnitude of the applied current being directly proportional to the iR drop (see Fig. 2g  and h).Notably, the MXene//Zn cells in 3 M ZnCl 2 solution (orange curves in Fig. 2g and h) featured the highest capacity at different rate, 179 and 63 mAh g − 1 , respectively, at current densities of 0.5 and 5 A g − 1 .However, after a few cycles (see Fig. S2), the performance declined sharply, as motivated above.The performance in 3 M ZnI 2 solution showed a reasonably capacity and a good rate performance, 135 and mAh g − 1 , respectively, at current densities of 0.5 and 5 A g − 1 .The upper cut-off potential window was limited to 1.1 V, as explained above, though the cells displayed a stable electrochemical performance during 10,000 cycles at a current density of 10 A g − 1 (see Fig. S4).The MXene// Zn cells in 3 M Zn(TFSI) 2 solution displayed a high capacity (136 mAh g − 1 ) at low rate (0.5 A g − 1 ), but at high rate the capacity decreased sharply (30 mAh g − 1 at 5 A g − 1 ), likely due to the bulky size of the TFSI − anion.The electrochemical behavior in 3 M ZnSO 4 showed a comparatively low discharge capacity of about 85 and 33 mAh g − 1 , at a current density of 0.5 and 5 A g − 1 , respectively.The lower capacity in ZnSO solution may be from spontaneous formation of Zn(OH) 2 /ZnSO 4 flakes on the surface of the MXene and Zn electrodes [46].The long-term cycling of the MXene//Zn cells in ZnSO 4 solution, however, displayed a stable behavior over 10,000 cycles (see Fig. S5).Notably, the MXene// Zn cells with Zn(CF 3 SO 3 ) 2 electrolyte showed a good rate performance (120 and 47 mAh g − 1 , at current densities of 0.5 and 5 A g − 1 ).A previous report described the performance of MXene//Zn cells in Zn(CF 3 SO 3 ) electrolyte in more detail [36].The following discussion focuses on suggesting solutions for increasing the energy density of the MXene//Zn devices by expanding the cycling potential window or by addressing the passivation problems observed in the electrodes with higher capacities, such as ZnCl 2 .

Expanding the operating potential of MXene//Zn cells using nonaqueous acetonitrile-based electrolytes
Generally, the use of a nonaqueous solvent offer a wider potential window in which the energy storage devices are cycled, which in turn   result in an increased device energy density due to a direct proportionality between the energy density and the operating potential.It has been reported recently that acetonitrile (AN) based electrolytes with Zn (CF 3 SO 3 ) 2 or Zn(TFSI) 2 salts are stable electrolytes for nonaqueous Zn batteries [16,[63][64][65].
Notably, when the MXene//Zn cells were cycled in 1.2 M Zn(TFSI) 2 / AN [66] and 1 M Zn(CF 3 SO 3 ) 2 /AN the upper cut-off potential was expanded to about 1.8 and 1.6 V, respectively (see Fig. 3).The MXene// Zn cells in 1.2 M Zn(TFSI) 2 /AN delivered capacities of about 32, 26, 16, 10, and 4 mAh g − 1 at current densities of 0.5, 1, 3, 5, and 10 A g − 1 , respectively (see Fig. 3a).Analogous capacities were obtained in 1 M Zn (CF 3 SO 3 ) 2 /AN of about 23, 18, 10, 7, and 3 mAh g − 1 (see Fig. 3c).A comparison between the electrochemical behavior in aqueous and nonaqueous electrolytes in the presence of the same anion carriers is demonstrated in Figs.3b and d.The results clearly show that the nonaqueous electrolytes offer a wider potential window, though the specific capacity decreases about 3-6 times as compared to aqueous electrolytes (see Table 1).The capacity loss observed in the nonaqueous electrolyte can be attributed to 1) the lower ionic conductivity of the nonaqueous electrolyte, 2) the lower concentration of the Zn salts in nonaqueous medium due to limited solubility, and (3) the absence of H + intercalation which can facilitate more access to active sites than the Zn 2+ ions [28,53,54].Therefore, nonaqueous electrolytes are promising candidates for expanding the operating voltage of the MXene//Zn cells, however, the nonaqueous solvent result in much lower capacity and poor rate performance.Accordingly, modifications need to be introduced for the MXene electrode manufacturing in order to obtain higher specific capacities than those reported in this study.It should be pointed out that the use of water-in-salt (WiS) electrolytes can also expand the potential window of electrochemical cycling by suppressing the potential of oxygen evolution [29].However, the use of 1 m Zn(CF 3 SO 3 ) 2 + 21 m LiTFSI (not shown) did not enable the expansion of the potential window owing to the oxidation of the MXene electrodes, resulting in passivation and failure of the cells after a few cycles using an upper cutoff potential higher than 1.3 V.An alternative route to increase the energy density of the MXene//Zn devices is to address the problems observed in some aqueous electrolytes, such as ZnCl 2 solutions in which the highest capacity was observed.

Engineering the Zn electrode surface using Ti 3 C 2 T z MXene coatings 2.4.1. Suppressing the passivation in MXene//Zn cells with ZnCl 2 electrolyte
Recalling the electrochemical behavior of the mixed MXene//Zn cells in 3 M ZnCl 2 , it showed a good initial performance, though a fast fading was observed after a few cycles at different rates (see Fig. 4a-c).For example, the initial capacity at 3 A g − 1 was about 96.6 mAh g − 1 , while the capacity dropped to about 19.8 mAh g − 1 during long term cycling due to passivation of the Zn electrode surface [48] as a result of the formation of oxide and/or hydroxide side deposits (see Fig. 4a-c and black line in Fig. 4g).The XRD pattern of the used Zn electrode showed an emerging peak originating from the surface deposits after cycling in 3 M ZnCl 2 (see Fig. S6).The EIS measurements clearly confirmed the passivation of the surface after a few cycles at different rate (see Nyquist plots in Fig. 4c).One possible solution to address this passivation problem is the use of highly concentrated ZnCl 2 solution [48].
Notably, the mixed MXene//Zn cells with 15 M ZnCl 2 electrolyte solution featured an improved long-term cycling stability as compared to that with 3 M ZnCl 2 (see Fig. 4d-f and blue line in Fig. 4g).For instance, the capacity retention in 3 M ZnCl 2 was 21% after 2000 cycles at a current density 3 A g − 1 , whereas the corresponding capacity retention in 15 M ZnCl 2 was 68%.This enhancement can be attributed to the formation of dense and smooth plated Zn as compared to the fluffy plated Zn formed in 3 M ZnCl 2 [48], in addition to the inhibition of side reactions that lead to formation of side products Zn(OH) 2 and ZnO in the presence of the highly concentrated ZnCl 2 [48].The EIS measurements confirmed the absence of surface passivation after a few cycles at different rate (see Nyquist plots in Fig. 4f).However, the MXene//Zn cells with 15 M ZnCl 2 solution failed after about 2000 cycles, which can be attributed to a short circuit as a result of Zn dendrite growth.
As mentioned earlier, the use of a metallic Zn negative electrode has drawbacks, and surface coatings [13,37,38] based on different materials, including Ti 3 C 2 T z MXene [42], were previously employed to address these challenges in a ZnSO 4 electrolyte.Analogously, we here examined the effect of a Ti 3 C 2 T z surface coating on the operation of MXene//Zn cells with Cl − anion carriers.Fig. 4h shows a schematic illustration of the MXene//Ti 3 C 2 T z -coated Zn cells, and Fig. 4i displays the SEM cross-section for the Ti 3 C 2 T z -coated Zn electrodes.The XRD pattern of the Ti 3 C 2 T z -coated Zn was analogous to that of the pristine Zn electrodes except the presence of the characteristic low angle peak associated with the Ti 3 C 2 T z MXene (see Fig. S6).As can be seen in Fig. 4i, a smooth few micrometer thick layer of Ti 3 C 2 T z was formed on the Zn surface.Notably, the presence of Ti 3 C 2 T z MXene layer on the surface of the Zn foil significantly improved the long-term cyclability in 15 M ZnCl 2 electrolyte, and the cells could be reversibly cycled over 10,000 cycles with a capacity retention of about 126% (see Fig. 4j-l and red line in Fig. 4g).The latter observation can be assigned to the increased efficiency of Zn plating-stripping, the inhibition of dendrite growth on the surface, as well as the prevention of side reactions occurring on the surface of a bare Zn foil, which in turn prevents passivation of the surface.The XRD pattern of the Ti 3 C 2 T z -coated Zn after cycling in 15 M ZnCl 2 was analogous to that before cycling (see Fig. S6).The Nyquist plots in Fig. 4l clearly show the absence of surface passivation.It should be noted that the morphology of the MXene electrode was maintained after long-term cycling, and that the Ti 3 C 2 T zcoated Zn showed a smooth morphology for the plated Zn on the surface (see Fig. S7).
Fig. 5a shows images of the different pristine and coated Zn electrodes before and after cycling in ZnCl 2 solutions, where it can be clearly seen that deposits were not observable on the Ti 3 C 2 T z -coated-Zn electrodes after cycling.Fig. 5b and c show schematic illustrations of the surface of a bare Zn electrode and the Ti 3 C 2 T z -coated-Zn electrode, respectively.
The use of thinner MXene electrodes can enhance the rate performance of the MXene//Zn cells, and inspired by the results of Ti 3 C 2 T zcoated Zn in 15 M ZnCl 2 solution, we examined the performance of 2.5 μm thick MXene electrodes (see Fig. 5d and Fig. S8).The thin-MXene// Ti 3 C 2 T z -coated Zn cells delivered capacities of about 200, 71, and mAh g − 1 at scan rates of 0.5, 100, and 200 mV s − 1 , respectively.Moreover, the variation of energy density with the power density (Ragone plot) of the thin-MXene//Ti 3 C 2 T z -coated Zn cells showed superior performance compared to the MXene results reported previously in the literature (see Fig. 5e).The energy density was about 128 and Wh kg − 1 at a power density of 0.179 and 19.8 kW kg − 1 , respectively.
Ref. [29] Ref. [25] Ref [30] Ref. [36] Zn foil MXene coated Zn foil (e) Variation of the energy density with power density for the thin-MXene//Zn cells in 15 M ZnCl 2 , determined from cyclic voltammetry, compared to the published literature for MXene based electrodes; blue squares this work, green asterisk Ref [30], black circles Ref. [25], red diamonds Ref. [36] and purple triangles Ref. [29].(For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Improving the coulombic efficiency at low rate
The use of Ti 3 C 2 T z MXene as a surface coating layer to improve the plating-stripping of MXene//Zn cells was also tested in 3 M Zn(CF 3 SO 3 ) 2 solution.Notably, the presence of a Ti 3 C 2 T z layer on the Zn surface improved the reversibility without affecting the accessible capacity of the MXene//Zn cells.At different rates the accessible capacity is similar to or even better than that in cells with an uncoated Zn foil (see Fig. 6 and Fig. S9).The coulombic efficiency clearly correlates with the improvement of the reversibility of MXene//Ti 3 C 2 T z -coated Zn (see Fig. 6e and f).This enhancement in the cell reversibility can be attributed to the superior reversibility of Zn plating-stripping on Ti 3 C 2 T zcoated Zn compared to the bare Zn foil.The EIS measurements displayed in Fig. 6g and h showed that the charge transfer resistance as well as the internal resistance of the MXene//Ti 3 C 2 T z -coated-Zn cells are lower than those of the MXene//Zn cells, which can be attributed to the inhibition of side product formation as briefly explained above.The CV before and after long-term cycling showed analogous behavior, indicating the high reversibility of the cells (see Fig. S10).

Conclusions
In summary, freestanding Mo 1.33 CT z -Ti 3 C 2 T z composite MXene films of a weight ratio of 3 Mo 1.33 CT z to 1 Ti 3 C 2 T z, were prepared using a onestep protocol.The effect of anion carriers (Cl − , I − , SO expansion of the upper cut-off potential up to1.8 V; however, the accessible capacity was reduced 3-6 times as compared to the use of an aqueous electrolyte, most likely due to the low ionic conductivity as well as absence of proton intercalation.The use of 15 M ZnCl 2 improved the cyclability of the MXene//Zn cells, though a short circuit was observed after 2000 cycles due dendrite growth. Engineering the surface of the Zn foil with addition of a fewmicrometer thick Ti 3 C 2 T z MXene significantly prevented the passivation associated with ZnCl 2 solutions and enabled long-term cyclability (over 10,000 cycles) in 15 M ZnCl 2 electrolyte.The MXene//Ti 3 C 2 T zcoated-Zn cells featured high capacity (200 mAh g − 1 at a scan rate of 0.5 mV s − 1 ) and an outstanding rate capability (36% capacity retention at a scan rate of 100 mV s − 1 ).EIS, XRD, and SEM confirmed that the MXene layer increased the efficiency of Zn plating-stripping, suppressed the dendrite growth, and prevented the side reactions which form oxide and/or hydroxide deposits.Furthermore, the use of Ti 3 C 2 T z MXene in 3 M Zn(CF 3 SO 3 ) 2 solutions improved the coulombic efficiency of the MXene//Ti 3 C 2 T z -coated Zn cells without affecting neither the accessible capacity nor the rate capability.All in all, the MXene can offer a dual function in ZHSC.It can be used as positive electrode material delivering a capacity up to 200 mAh g − 1 , which is among the highest previously reported capacity for MXene electrodes.Furthermore, MXene can be employed as a surface coating layer for the Zn foil negative electrode, enabling the use of low-cost ZnCl 2 solutions.The latter is important for future applicability within the field of energy storage.

Synthesis of Mo 1.33 CT z MXene
To prepare Mo 1.33 CT z , its precursor, (Mo 2/3 Sc 1/3 ) 2 AlC was synthesized as reported previously in Ref. [51], via solid-liquid reaction of elemental powders of Mo (3 to 7 μm, Alfa Aesar, Kandel, Germany, 99.999 wt% purity), Sc (99.99%,Stanford Advanced Material, USA), Al (99.5%, 325 mesh, Alfa Aesar, Kandel, Germany) and C (99.9995%, 200 mesh, Alfa Aesar, Karlsruhe, Germany).The powders Mo, Sc, Al and C were mixed in a molar ratio of 4/3:2/3:1:1 respectively.The powders were mixed manually using an agate mortar and pestle (to ensure a homogenous mixing, the total mass of the powders should not exceed 5 g in one turn).The mixture was transferred to an alumina crucible and placed in a horizontal furnace.The furnace was heated to 1550 • C and held for 24 h under 5 sccm Ar flow.Both the heating and cooling were set at 3 • C/min.The lightly sintered product was crushed using a mortar and pestle and sieved through a 450-mesh sieve.
To selectively etch both Al and Sc, 2 g of (Mo 2/3 Sc 1/3 ) 2 AlC were put slowly, 0.5 g each at a time with 2 min waiting time between each 0.5 g, in a Nalgene bottle containing 40 ml of ≈ 50% hydrofluoric acid, HF, (Sigma Aldrich, Sweden).The mixture was then stirred using a Teflon coated magnetic stirrer at room temperature for 30 h.Afterwards, the mixture was washed using deionized (DI) water, seven cycles of washing were performed each of 40 ml of water, and in each cycle MXene and water were added to a centrifuge tube, then hand shaken for 30 s, then centrifuged at 6000 rpm for 1 min, then the supernatant was decanted.After the 7th cycle, the pH of the supernatant was checked to ensure that it is between 6 and 7.For delamination of the multilayered powder, 10 ml of 54-56 wt% tetrabutylammonium hydroxide, TBAOH, (Sigma Aldrich, Sweden) solution was added to the powder in a centrifuge tube and shaken for 5 min using Corning LSE Vortex Mixer at a speed of 1700 rpm.Afterward, the TBAOH was decantated after centrifugation for 3 min at 6000 rpm.The sedimented powder was washed three times with 40 ml 100% ethanol, hand shaken for 30 s, and decanted again using centrifugation (5 min at 600 rpm).The powder was finally washed three times with 40 ml DI water, in this step no shaking was done to prevent spontaneous delamination.The delamination was done by adding 30 ml of DI water and shaken for 5 min using the Vortex mixer at 1700 rpm.The obtained mixture was centrifuged for 1 h at 2500 rpm to obtain a colloidal suspension of single to few flakes of Mo 1.33 CT z MXene.

Synthesis of Ti 3 C 2 T z (MXene)
The Ti 3 AlC 2 MAX phase was synthetized using solid-liquid reaction of TiC (Alfa Aesar, 98 + %), Ti (Alfa Aesar, 98 + %) and Al (Alfa Aesar, 98 + %) of 1:1:2 molar ratio of 2:1.25:2.2,respectively as described previously [67].The mixture was placed in an alumina crucible and inserted in a horizontal furnace.The furnace was heated to 1380 • C with a rate of 3 • C/min under 5 sccm Ar flow.The heating temperature was kept at 1380 • C for 2 h and then cooled down to room temperature at a rate of 3 • C/min.The lightly sintered product was crushed using a mortar and pestle and sieved through a 450-mesh sieve.To remove the excess Al, 3 g of the powder was added to 40 ml of 12 M hydrochloric acid, HCl, (Fisher, Technical grade) and stirred using Teflon magnetic stirrer for 24 h (care was taken when adding the powder to HCl to avoid aggressive reaction, 0.5 g was added each time and left to react for 3 min before adding another 0.5 g).The mixture was washed with DI water for three cycles each of 40 ml, each time the mixture was centrifuged at 6000 rpm for 1 min followed by decanting of the supernatant.After washing the final mixture was filtered using vacuum filtration then the powder was left to dry in air overnight.
One gram of the dried powder was added to a mixture of 12 ml HCl, 6 ml H 2 O, and 2 ml HF and left under stirring for 24 h at 35 • C according to Ref. [68] Afterward, the mixture was washed with water as described above in the section of etching of (Mo 2/3 Sc 1/3 ) 2 AlC.After washing, the final mixture was vacuum filtered to obtain the multilayered powder where 1 g was then added to a LiCl solution (6 g of LiCl (Alfa Aesar, 98 + %) dissolved in 25 ml of DI water) and stirred for 24 h at room temperature.For delamination, the mixture was washed with DI water through several cycles each of 40 ml of water, the washing was stopped once a black supernatant was observed after centrifugation at 6000 rpm for 1 min.After that 40 ml of water were added to the sediment and was shaken for 5 min using the Vortex shaker at 1700 rpm followed by centrifugation for 1 h at 2500 rpm to obtain a supernatant of colloidal suspension of single to few layers of Ti 3 C 2 T z .

Preparation of composite MXene films
The composite MXene films were prepared using straightforward one step protocol.In a typical experiment, the pristine MXene suspensions were mixed in a weight ratio 3 Mo 1.33 CT z : 1 Ti 3 C 2 T z .The mixed suspensions were agitated for 30 s and then filtrated under vacuum using Celgard 3501 membrane.The as prepared films were stored in Ar filled glove box.The active mass loadings ranged between 0.5 and 2.1 mg cm − 2 .

Preparation of Ti 3 C 2 T z -coated-Zn
The Ti 3 C 2 T z suspension was diluted to about 1 mg ml − 1 , and then the suspension was drop cast into freshly polish Zn foil disks to cover the surface of the foil, roughly about 25-40 μl of the suspension per 4 mm Zn foil disk, then allowed to dry in ambient at room temperature, and then stored in Ar filled glove box for further use.

Materials characterization and electrochemical measurements
Scanning electron microscope (SEM, LEO 1550 Gemini) was used to investigate the morphology of the samples and their thickness.The structure of the composite MXene was determined by X-ray diffraction (XRD) using PANalytical diffractometer (Cu K α radiation λ = 1.54 Å, step size 0.0084 • , time per step 20 s).

Fig. 1 .
Fig. 1.Synthesis and morphology of the mixed MXene electrodes: (a) Schematic illustration of the synthesis of an ideal MXene composite material from individual MXene suspensions.(b), (c), (d) SEM cross-section images of the mixed MXene films at different magnifications.The inset in (b) displays a photograph of a flexible freestanding film.

Fig. 2 .
Fig. 2. Effect of anion carriers on the electrochemical performance of MXene//Zn cells: (a) and (b) schematic illustrations of the MXene//Zn cells in the charging and discharging states, respectively.(c) Variation of the potential of the Zn electrode versus the Ag/AgCl reference electrode during the operation of the MXene//Zn cells.(d) and (e) CVs of the MXene//Zn cells in different electrolytes at scan rates of 0.5 and 10 mV s − 1 , respectively.(g) and (h) Charge-discharge profiles of the MXene// Zn cells in different electrolytes at current densities of 0.5 and 5 A g − 1 , respectively.(f) and (i) Nyquist plots in different electrolytes before and after the precycling step, respectively.

Fig. 4 .
Fig. 4. Engineering the surface of a Zn electrode using Ti 3 C 2 T z MXene coatings: (a), (b), and (c) CVs at a scan rate of 10 mV s − 1 , charge-discharge profiles, and Nyquist plots, respectively, for MXene//Zn cells in 3 M ZnCl 2 solution.(d), (e), and (f) CVs at a scan rate of 10 mV s − 1 , charge-discharge profiles, and Nyquist plots, respectively, for MXene//Zn cells in 15 M ZnCl 2 solution.(g) Long-term cycling of the MXene//Zn cells.It should be noted that the mixed MXene//Zn cells were cycled for a few cycles at different rates prior to the long-term cycling test.(h) Schematic illustration of the MXene//Ti 3 C 2 T z -coated-Zn cells.(i) SEM cross-section images of the Ti 3 C 2 T z -coated Zn, the right panel showing a 90 • tilted high magnification SEM image of the Ti 3 C 2 T z .(j), (k), and (l) CVs at a scan rate of 10 mV s − 1 , charge-discharge profiles, and Nyquist plots, respectively, for MXene//Ti 3 C 2 T z -coated Zn cells in 15 M ZnCl 2 solution.

Fig. 5 .
Fig. 5. (a) Images of the pristine and coated Zn electrodes before and after cycling in ZnCl 2 solutions.(b) and (c) Schematic illustrations of the surface of a bare Zn electrode and the Ti 3 C 2 T z -coated Zn electrode, respectively.(d) Variation of the discharge capacities and coulombic efficiency of the thin-MXene//Zn cells in 15 M ZnCl 2 .(e) Variation of the energy density with power density for the thin-MXene//Zn cells in 15 M ZnCl 2 , determined from cyclic voltammetry, compared to the published literature for MXene based electrodes; blue squares this work, green asterisk Ref[30], black circles Ref.[25], red diamonds Ref.[36] and purple triangles Ref.[29].(For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

4 2 −Fig. 6 .
Fig.6.Engineering the surface of a Zn electrode using a Ti 3 C 2 T z MXene coating in 3 M Zn(CF 3 SO 3 ) 2 : (a) and (b) CVs at a scan rate of 0.5 mV s − 1 and galvanostatic charge-discharge plots at a current density of 1 A g − 1 , respectively, for the MXene//Zn cells (grey lines) and MXene//Ti 3 C 2 T z -coated-Zn cells (blue lines).(c) and (d) CVs at different scan rates and charge-discharge profiles at different current densities, respectively.(e) and (f) Variation of chargedischarge capacity and coulombic efficiency with scan rate and current density, respectively, for the MXene//Zn cells (grey diamonds and circles) and MXene//Ti 3 C 2 T z -coated-Zn cells (blue triangles and crosses).(g) and (h) comparison of Nyquist plots before and after precycling, respectively, for MXene//Zn cells (grey lines) and MXene//Ti 3 C 2 T zcoated-Zn cells (blue lines).(For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) ZnCl 2 enabled the reversible use of ZnCl 2 electrolyte in mixed MXene//Zn cells.Furthermore, the life span as well as the reversibility of the mixed MXene//Zn cells was significantly improved after coating the surface of the Zn foil with Ti 3 C 2 T z MXene.The MXene//Ti 3 C 2 T zcoated-Zn cells can deliver a reversible capacity up to 200 mAh g − 1 and a capacity retention of about 36% at high rate.
−, and TFSI − ) on the accessible capacity as well as the lifespan and rate capability of MXene//Zn hybrid supercapacitors.The use of acetonitrile-based nonaqueous electrolytes is also investigated, to expand the operating potential window.The use of high concentration, 15 M,

Table 1
Summary of the electrochemical performance of various MXene//Zn cells in different electrolytes.