Robust Biocompatible Fibers from Silk Fibroin Coated MXene Sheets

Conductive fibers are needed for the development of flexible electronic and biomedical devices. MXene fibers show great promise for use in such applications because of their high conductivity. Current literature on MXene fiber development highlights the need for improving their mechanical properties and investigation of biocompatibility. Here the use of silk fibroin biopolymer as a MXene formulation additive for the production of MXene fibers is studied. It is found that the favorable silk fibroin–MXene interactions resulted in improved durability, withstanding up to 1 h of high frequency sonication in buffered solutions. Furthermore, fibers with ≈5 wt% silk fibroin displays interesting properties including high conductivity (≈3700 S cm−1), high volumetric capacitance (≈910 F cm−3), and non‐cytotoxicity toward THP‐1 monocytic cells. The results presented here provide an important insight into potential use of MXene fibers in flexible electronics and biomedical applications.

tensile strengths up to ≈570 MPa have been reported (e.g., MXene-CMC-borate films). [23] However, unlike their film counterparts, it remains challenging to fabricate durable fibers with high MXene loading due to their insufficient inter-sheet connectivity, [20,27] since most MXene dispersions are produced with relatively low average lateral sheet size, <1 µm. [18,19,27] Although some studies have successfully produced MXene sheets (>3 µm) [28,29] resulting in MXene fibers with tensile strengths ranging from ≈100 to 350 MPa, [30,31] the defect-induced lack of flexibility and extensibility in these fibers is still prominent (<1% strain). In this regard, bioinspired strategies could be adapted to address these aforementioned issues, as these strategies are already known to provide apparent sheet size increase resulting in flexibility and fatigue resistance in non-fiber MXene architectures. [13,23,24] However, it is yet unknown if these strategies can be applied on MXene fiber production. In addition, while recently developed bioinspired strategies showed overwhelming success in providing enhanced tensile strengths for non-fiber MXene architectures, their potential cytotoxic effects, which are a crucial criterion for many applications, also remains relatively unknown and unexplored. As of 2021, only 20 out of more than 120 articles reporting MXenes in biomedical applications were found to present any biocompatibility assessment. [32] Due to the dependence of MXene biocompatibility on structure and synthetic approach, [32] such limited information inhibits their development as practical implantable devices.
In this study, we investigate the fabrication of a robust MXene composite fiber in order to provide critical information on how to address the setbacks faced by MXene-based fibers toward their development as wearable devices. Using a strategy of co-assembling MXene with silk fibroin as a bio-derived filler, we show that mixtures with ≈5 wt% silk content can produce tough (≈0.7 MJ m −3 ) fibers, with high conductivity (≈3700 S cm −1 ) and excellent volumetric capacitance (≈910 F cm −3 ). These fibers also showed delamination resistance under agitation forces such as ultrasonication, and conductivity retention upon prolonged storage in simulated physiological conditions (phosphate buffered saline [PBS] pH = 7.4 at 40 °C). More importantly, MXene-silk fibers demonstrated high cytocompatibility (cell viability up to ≈200%) in a THP-1 monocytic cell culture, which we believe to be a crucial first step toward developing MXene fibers for their practical use in advanced applications.

MXene, Silk, and MXene-Silk Dispersions
First, single-layer MXene sheets were produced by selectively etching aluminum from Ti 3 AlC 2 MAX phase using a minimally intensive layer delamination (MILD) method as described in previous works. [13,14,28,29,33,34] Silk fibroin solution was extracted by first degumming raw silk fibers for 180 min at 98 °C with 3 g L −1 sodium carbonate in a silk (g) to liquor (mL) ratio of 1:50. Degummed fibers were then dissolved for 300 min at 60 °C with 9.3 m lithium bromide, before being dialyzed in milli-Q water for three days to remove the lithium salt. [35] Additional details of the MXene synthesis and silk fibroin extraction, with their respective full characterization data are summarized in the Supporting Information. Briefly, our scanning electron microscope (SEM), atomic force microscope (AFM) and X-ray diffraction (XRD) data (Figure 1a and Figure S1b,c, Supporting Information) confirmed successful etching and exfoliation, resulting in unilamellar MXene sheets that were ≈2 nm thick and ≈0.5 to 1.5 µm across. Meanwhile, the size and morphology of silk fibroin was also characterized using AFM and found to be rice-shaped aggregates with diameters of ≈1-2 nm and lengths of ≈6-10 nm (Figure 1a).
The effect of silk fibroin on MXene dispersion properties such as colloidal stability, viscosity, viscoelastic behavior, and liquid crystallinity was first studied. Using dynamic light scattering (DLS), the addition of small amounts of silk (≈5 wt%) to MXene dispersions resulted in an apparent increase in MXene's average hydrodynamic diameter from ≈700 nm to ≈2.0 µm (Figure 1b). Likewise, the peak at ≈100 nm, corresponding to small MXene fragments also shifted to ≈250 nm. Furthermore, the peaks at ≈2, 10, 80 and 500 nm, corresponding to silk fibroin and related aggregates, were all absent or significantly reduced in the mixed dispersion, suggesting their adsorption onto the MXene surface ( Figure 1b). This implies strong mutual attraction between the silk and MXene, which was confirmed with AFM, where fibroin was visible across the MXene sheet surface (Figure 1a and Figure S2, Supporting Information). However, it should be noted that agitation after silk addition to MXene may result in some breakup of large sheets. [36] This was observed in the DLS with an intensity decrease of the peak representing large sheets, accompanied with an increase in intensity for the peak of small sheets for the silk-containing sample ( Figure 1b). In addition, due the 2D anisotropic geometry of MXene, the large sheets may also fold or rotate contributing to the apparent increase in the number of the small sheets. Nonetheless, the overall position of these DLS peaks shifted toward a larger size range, indicating a bridging-induced apparent increase in lateral sizes. Despite the addition of a relatively more neutral and typically aggregating silk fibroin (ζ-potential ≈ −10 mV), [37,38] the mixed dispersion retained colloidal stability close to that of pristine MXene (approximately −35 and −43 mV, respectively) ( Figure 1c). To further elucidate the silk-MXene interactions, the rheology of the mixed dispersions were evaluated. Although all dispersions exhibited shear thinning behavior (Figure 1d), the zero-shear viscosity (η0) of MXene-silk mixture was found to be ≈310 Pa s, which was much higher than that of silk (0.38 Pa s) and slightly higher than MXene (250 Pa s). The oscillatory sweeps revealed both MXene and MXene silk dispersions exhibited gel-like behavior for the whole angular frequency range (G′/G″ > 1, Figure 1e). In addition, MXene-silk exhibited higher G′/G″ ratio than MXene, indicating silk-induced improved viscoelasticity (Figure 1e), which is an important requirement to achieving MXene spinnability. [28,39,40] Finally, a visual assessment of birefringence using polarized optical microscopy (POM) confirmed retention of liquid crystallinity, despite the addition of non-liquid crystalline silk (Figure 1f). The appearance of dark brushes, also known as Schlieren textures, indicate preferential basal plane alignment of MXene sheets, [14,28,41] a property facilitating MXene assembly into architectures such as film casting [29] and wet-spinning [14,28] which involve the application of shear forces. Schlieren textures were fully retained even when silk content was increased to 10 wt% (M 90 ).

Nature of MXene-Silk Interactions
To gain more insight into MXene and silk interactions, chemical analysis via X-ray photoelectron spectroscopy (XPS) was carried out. The appearance of a peak in the ≈400 eV region of MXene-silk survey spectra (Figure 2a), indicates the presence of nitrogen-containing silk fibroin in the composite. [33,42] However, its N 1s core spectra (Figure 2b) revealed that all nitrogen species in the composite exists purely as amide groups. [43] Despite increased contributions for CN (≈286.6 eV), CO (≈287.1 eV) and CO (≈288.7 eV) peaks in MXene-silk C 1s core spectra (Figure 2c) which correspond to the presence amino acid functional groups from silk, [43] no additional Ti-coordination in the Ti 2p core spectra was observed ( Figure 2d). This suggests that MXene-silk interactions were electrostatic or h-bonding mediated, [13,24,25]  www.advmatinterfaces.de particularly as silk fibroin is known to be dominated by repeating glycine, alanine and serine residues that form tightly packed h-bonded beta sheet nanocrystals. [44] However, we observed significantly increased CTiOH (≈532.6 eV) peak intensity in MXene-silk's O 1s spectra (Figure 2e), indicative of higher amounts of OH functionality in the sample compared to pristine MXene. The additional OH character could be a result of filling remaining Ti vacancies when OHrich silk fibroins were introduced on the MXene surface. It could also be due to substitution of F-terminations as these groups are known to be easily exchanged with OH when exposed in aqueous conditions, as evident on the lower F 1s intensity in MXene-silk survey spectra (Figure 2a). [45] The abundance of OH functionality, which also provides additional H-bonding sites, is believed to drive the strong interactions between MXene and silk, as observed in previous sections. [43] We also note the appearance of a small HOH peak in the O 1s spectra of pristine MXene (Figure 2e). This peak was ascribed to water intercalation in MXene. This peak was not present in the O 1s core spectra of MXene-silk, implying that a coating of silk inhibits water intercalation in MXene. However, for mechanistic elucidation of this effect a more in-depth investigation is required.

Wet-Spinning of MXene-Silk Composite Fibers
Composite fibers of increasing silk content from 0 to 15 wt% (M 100 , M 97.5 , M 95 , M 92.5 , and M 85 , respectively) were formed by extruding the MXene-silk dispersions into a coagulation bath containing 0.5 wt% chitosan in 1% v/v acetic acid solution (Figure 3a) using a small-scale wet-spinning approach described in our previous works. [14,28] The fibers possess irregular cross-sections, with diameter ranging between ≈20 to 50 µm (Figure 3b), with MXene sheets highly aligned parallel to the fiber axis ( Figure 3b). When the silk content increased, the diameter increased with lesser voids but is more corrugated in shape (Figure 3b), which we believe was influenced by varying the MXene-silk weight ratio. However, previous studies have demonstrated enhanced MXene fiber morphology by increasing the spinning dope concentration. [14] This approach, as well as the optimization of other spinning parameters such as shear rate, coagulation bath composition, and drying conditions to further improve fiber morphology may be explored in future studies. [28,35,46]

MXene Alignment and Interlayer (d)-Spacing
To gain insights about the structural changes observed from the SEM images, small-angle X-ray scattering (SAXS)/wideangle X-ray scattering (WAXS) measurements were carried out (Figure 4a). The resulting 2D SAXS/WAXS scattering patterns of all MXene fiber samples exhibited scatterings at q ≈ 0.5 Å −1 and q ≈ 1.0 Å −1 , attributed to the (002) and (004) signals of MXene, respectively ( Figure 4b). [28,29,47] We observed that starting with M 95 fibers, MXene's (002) signal starts to split, with one remaining at ≈0.5 Å −1 and the other shifted toward a lower q value of ≈0.27, indicating the presence of MXene sheets with different interlayer (d)-spacing ( Figure 4b). Numerically integrating these signals with respect to q y revealed that these splitting corresponds to d-spacings of ≈1.23 and ≈2.2 nm, respectively ( Figure 4c). The detection of MXene with ≈2.2 nm spacing indicates silk fibroin intercalations in between the sheet layers, which according to AFM, also have an approximate diameter of ≈1-2 nm (Figure 1a). The splitting of the signal into 2 distinct peaks is believed to be caused by the incomplete coverage of silk fibroin over the MXene sheets as also shown by AFM (Figure 1a), therefore some fibroin-free sections remained to manifest interlayer spacing values similar to pristine MXene

www.advmatinterfaces.de
The degree of MXene sheet alignment in the fibers was also assessed quantitatively using their respective Herman's orientation factor (f). This was acquired through integration of the (002) peaks with respect to the azimuthal direction (Figure 4d), where MXene sheet alignment with respect to the q y direction is indicated by f values closer to 1. [28,29,35] We found that insertion of silk in between MXene sheets resulted in decreased alignment from an f value of ≈0.842 for pure fibers, down to ≈0.686 for the more corrugated M 90 fibers (Figure 4d). To understand the implications of these observed changes in MXene alignment and spacing, the physical and electrochemical properties of the fibers were also assessed ( Figure 5).

Properties of Wet-Spun MXene Fibers
Overall, stress-strain curves illustrate that small amounts of inserted silk fibroin (≤2.5 wt%) enhance the tensile strength, however, it has minimal effect on the resulting fiber's tensile strain (Figure 5a). For example, fibers at 2.5 wt% silk (M 97.5 ) resulted in noticeable increase in tensile strength (≈80 MPa, Figure 5a), and toughness (≈0.45 MJ m −3 , Figure 5b) compared with pristine MXene fiber (M 100 ) at ≈60 MPa and ≈0.13 MJ m −3 , respectively. Further improvement in toughness (≈0.7 MJ m −3 ) was observed upon increasing the silk content to 5 wt% (M 95 ), while still retaining a significant fraction of electrical conductivity (≈3700 S cm −1 ) compared to pristine MXene (≈7500 S cm −1 ) (Figure 5c). These results imply that despite the slight misalignment observed from the SAXS/WAXS data, strong silk-induced sheet adhesions mainly drive the improvement in toughness, enough to circumvent the detrimental effects of uneven sheet spacing distribution, which often leads to weakened stress distribution in MXene fibers. [14,18,19] On the other hand, fibers with higher silk content up to 15 wt% silk (M 85 ) showed decreased tensile strength (≈35 MPa), but with much higher maximum strain (≈5.8%) and toughness (≈1.35 MJ m −3 ). These results suggest that "excessive" amounts of "soft" silk fibroin lowered the tensile strength of MXene fibers, however, the enhanced H-bond-driven adhesions between sheets, resulted in higher breaking strains, further increasing the overall toughness of the fibers.
Meanwhile, probing the electrochemical properties showed cyclic voltammetry (CV) curves for current collector-free MXene-silk fibers (M 95 and M 90 ) with noticeably lower area than that of M 100 (Figure 5d). These values correspond to volumetric capacitance values of ≈1200, ≈910 and 750 F cm −3 for M 100 , M 95 and M 90 fibers, respectively. Whilst both MXene-silk fibers exhibited lower volumetric capacitance than M 100 , the M 95 and M 90 fibers had higher capacitance retention at increasing scan rates (Figure 5e). Moreover, when the CV scan rate was increased from 2 to 500 mV s −1 , M 95 and M 90 fibers displayed a capacitance retention of ≈46% and ≈40% respectively, both relatively higher than that of pristine fibers (≈35%). These results show that despite silk being insulating, the increased MXene interlayer d-spacing facilitated electrolyte and ion transport into the fiber. [3,13,14,48] The M 95 fiber also showed stability over repeated charging and discharging, retaining ≈94% of its capacitance with ≈100% coulombic efficiency even after 10 000 CV cycles (Figure 5f), demonstrating potential high performance energy storage applications.

Delamination Resistance, Stability in Dispersion, and Cytocompatibility
Adding silk to MXene fiber formulations also provided essential properties such as delamination resistance (Figure 6a,b), stability in physiological conditions ( Figure 6c) and cytocompatibility (via cytotoxicity tests, Figure 6d), which are all vital characteristics for their development toward practical applications. While the tensile and electrical/electrochemical properties of M 95 fibers were comparable to other wet-spun MXene fibers (Table S2, Supporting Information), these fibers display both delamination resistance and prolonged shelf-stability, which have not been observed for any MXene-derived fiber in the literature.
Similar to M 100 , the addition of 2.5 wt% silk led to fiber disintegration after 30 min ultrasonication, however further increases in silk content (>5%) resulted in reduced fiber damage over 60 min (Figure 6a). As confirmed through SEM crosssections (Figure 6b), sonicating M 100 fibers for 30 min resulted in drastic misalignment of MXene sheets in contrast to M 95 fibers which remained tightly packed. This structure retention for M 95 fibers is suspected to be an effect of stronger inter-sheet adhesion through additional H-bonding sites introduced by the added silk fibroin bridges (Figure 2e), which prevented MXene delamination. [43] In addition to sheet delamination, one significant limitation of MXene-based materials is their chemical stability in environmental conditions, especially in water. [20,49] This can be addressed through storage in either an oxygen-free atmosphere, [19,20] organic solvents, [20] with antioxidants [33] or frozen. [49] However, the long-term stability of MXene-based biomaterials under ambient conditions, mimicking practical applications, remains a challenge. Here, we found that the stability of the fibers in such conditions (PBS buffer or air), monitored through changes in electrical resistance, noticeably improved for M 95 over its pristine counterpart. M 95 fibers soaked in PBS at 40 °C remained more conductive exhibiting only a ≈33% increase in electrical resistance compared to ≈59% for M 100 (Figure 6c). In addition to the structure and packing preservation observed in Figure 6b, this conductivity retention is suspected to be due to either the water-resistant silk preventing of fiber swelling [50] or acting as a protective agent slowing MXenes oxidation rate. [43] Lower resistance change for M 95 fiber over its pristine counterpart was also observed when stored in air and in PBS at room temperature, with ≈4% and ≈17% change for M95 against ≈5% and ≈24% for M 100 , respectively (Figure 6b). Moreover, long-term stability tests over 7 months at room temperature (in air) revealed that M 95 only exhibited ≈190% increase in electrical resistance, ≈5× lower than the resistance increase of Cytotoxicity evaluations using THP-1 cells in RPMI-1640 media (with 10 vol% fetal bovine serum) revealed that M 95 fibers are potentially biocompatible and have noticeably higher cell viability (≈180%) relative to M 100 (≈150%) (Figure 6c). Increase in cell viability of up to ≈200% was observed with increasing silk content (M 95 ), while non-cytotoxic in nature, [51] pure silk was found to have lower cell viability (≈80%) than pure MXene. Interestingly, a combination of the two materials resulted in a biocompatible composite, which did not show any detrimental effect to the material's overall cytocompatibility, and in fact even promoted cell proliferation (cell viability > 100%). However, we acknowledge that the cell proliferative nature induced by silk-MXene synergistic behavior, is not well understood and could be a collective effect of several factors, including formation of unknown by-products with potentially cell-growth-triggering properties upon cells exposure to MXene-silk. Further understanding of this phenomenon is outside the scope of this study however deserves a more thorough investigation in the future.

Conclusion
MXene-silk fibroin composite fibers were fabricated and their properties investigated including cytotoxicity and durability under high frequency sonication. We found that composite fibers possessed improved properties such as physical stability and conductivity retention in physiological conditions, which are generally lacking in most studies reporting bioinspired MXene-fortification methods. Particularly, by adjusting the amount of silk in the spinning formulations, the conductive composite fibers can be tuned to possess either high tensile strength (≈80 MPa at <5 wt% silk), or high tensile strain and toughness (≈5.8% and ≈1.35 MJ m −3 , respectively, at >15 wt% silk). In addition, the protective nature of silk (even at just 5% content) resulted in fibers with significant resistance to delamination (under sonication up to 1 h) and conductivity retention, even when stored in physiological conditions up to seven days. Our results suggest that these improvements in fiber properties were brought by silk's abundant H-bonding sites providing multiple points of interaction with MXene that lead to enhanced inter-sheet adhesion. The combination of MXene-silk also exhibited enhanced cell-viability of up to ≈200% at 10% silk, providing an important insight into the potential use of MXene fibers in biomedical applications. Although other nanomaterial fillers such as nanocellulose [46,52] and graphene oxide [27] were also reported to result in higher mechanical properties, biocompatibility varies with every material. Thus, subsequent cytotoxicity testing after material fabrication is highly recommended. We believe that this crucial information on how MXene fibers achieve a combination of robust physical properties, retention of electrical conductivity and electrochemical properties, with additional data on biocompatibility will provide a path toward their development into functional wearable devices.

Experimental Section
Synthesis of MXene: Single-layer Ti 3 C 2 T x MXene sheets were synthesized through the selective etching of aluminum from Ti 3 AlC 2 MAX phase using the minimally intensive layer delamination (MILD) method described in previous works. [20,47,53] The etching solution was prepared by adding 1.6 g lithium fluoride (LiF, 99%, Sigma-Aldrich Pty Ltd) to 20 mL of 9 m hydrochloric acid (HCl, 12 m, Sigma-Aldrich Pty Ltd) solution, followed by stirring for 5 min to fully dissolve the LiF. 1 g of Ti 3 AlC 2 powder (Carbon-Ukraine Ltd., particle size < 40 µm) was slowly added to the etchant at room temperature and stirred for 28 h. The acidic suspension was washed with deionized (DI) water until pH ≈ 6 via centrifugation at 3500 rpm (5 min per cycle) and decanting the supernatant after each cycle. Multilayer MXene was removed as sediment by centrifugation at 1500 rpm for 30 min. The dark green supernatant mainly containing unilamellar MXene was then concentrated at 7000 rpm for 30 min. The MXene dispersion concentration was measured by vacuum drying a specific volume of the colloidal solution (concentration 15.0 ± 0.2 mg mL −1 ).
Preparation Silk Solution: Silk fibroin dispersion used in this work was prepared based on the method described in the authors' previous work. [35] Briefly, raw silk fibers were degummed (type of the dyeing machine) at 98 °C with 3 g L −1 aqueous sodium carbonate for 180 min, using a silk (g) to liquor (mL) ratio of 1:50. The degummed fibers were then dissolved in 9.3 m LiBr solution (7.5 mL of LiBr solution per 1 g of silk fibers) at 60 °C for 300 min. The resulting silk solutions were then dialyzed against DI water using regenerated cellulose tubing (3.5 kDa MWCO) for 3 days to remove the salts and finally yield a ≈6% (w/v) silk dispersion. The exact concentration of the final silk solutions was then calculated gravimetrically by averaging four dried (60 °C for 3 h) aliquots (0.5 mL) of each solution. The molecular weight of silk solution used in this study was determined to be 33 kDa, using size exclusion HPLC as described previously. [35] MXene-Silk Dispersion Preparation: MXene-silk dispersions (referred to as MX) were prepared by adding varying volumes of a MXene (≈25 mg mL −1 ) into a fixed volume of a silk dispersion to produce mixed dispersions of varying MXene:silk weight ratios (M 100 , M 99 , M 97.5 , M 95 , M 90 , and M 85 ), but maintaining a constant total solid concentration of 25 mg mL −1 . All dispersions were hand shaken for 10 min to ensure homogeneity. The dispersions were then incubated under ambient conditions for at least 1 h before use.
Fiber Wet-Spinning: As-prepared MXene and MXene-silk dispersions were used as spinning dopes for wet-spinning. First, a coagulation bath was prepared containing 0.5 wt% chitosan, (molecular weight 310 000-375 000, Sigma-Aldrich, Australia) dissolved in 5 mL acetic acid (Sigma-Aldrich, Australia) and 500 mL milli-Q water (Figure 2a). Dispersions were extruded into a coagulation bath using a blunt-tip needle (25-gauge, inner radius of 130 µm) and a syringe pump set to a flow rate of 3.0 mL h −1 (shear rate ≈485 s −1 ). The shear rate (γ, s −1 ) during spinning was estimated using Equation (1): where Q is defined as injection flow rate (m 3 s −1 ) and R is the needle inner radius (m). All fibers were manually collected as-spun onto a spool without draw or post-processing (e.g., washing). Characterization of MXene, MXene-Silk Dispersions, and Fibers: Images of samples mounted vertically on aluminum stubs were taken using a Supra 55-VP field emission SEM (Zeiss, Germany) at an accelerating voltage of 3 keV. MXene sheet thickness was acquired using a MultiMode 8-HR AFM (Bruker, USA) set at a 0.5 Hz scan rate. AFM samples were made of drop casted dilute MXene dispersions (10 µg mL −1 ) on pretreated silicon wafers (soaked overnight in a solution of 30 vol% H 2 O 2 and 18 m H 2 SO 4 at a 1:5 volume ratio). The size distribution of dispersed MXene sheets in dispersion was estimated through DLS on a Nano ZS Zetasizer (Malvern Instruments, UK) using a 1 mL aqueous dispersion (10 µg mL −1 ) placed in a disposable plastic cuvette using an average of five measurements for each sample. The liquid crystal behavior of www.advmatinterfaces.de dispersions was observed by looking at their respective birefringence patterns under a polarized optical microscope (POM) (Nikon Eclipse 80i, Japan) equipped with a home-made cell ( Figure S3, Supporting Information). The rheological properties of the dispersions were measured using a rheometer (TA Instruments HR-3, USA) equipped with a parallel plate geometry (cone-shaped with angle: 2°, diameter: 40 mm). A volume of ≈600 µL of the dispersions was carefully loaded to the rheometer stage to prevent sample shearing. Shear-induced change in viscosity in the samples was measured at shear rates ranging from 0.01 to 1000 s −1 . In this experiment, the viscosity of the dispersions at 0.01 s −1 were considered as the zero-shear viscosity due to the difficulty in equilibrating dispersions in their undisturbed state. [39] The viscoelastic properties, specifically the elastic modulus (G′) and viscous modulus (G″) were also measured as a function of frequency (constant strain) at an amplitude of 0.1%.
Chemical analysis and binding energy profile were characterized on an RBD upgraded PHI-5000C ESCA system (Perkin Elmer) with Mg Kα radiation. Survey spectra were collected at an energy range of 0 to 1100 eV, pass energy of 160 eV, and a resolution of 1 eV. High resolution spectra of C 1s, O 1s, and Ti 2p were collected at a pass energy of 20 eV and a step size of 0.1 eV. The X-ray photoelectron spectroscopy (XPS) data were analyzed using CasaXPS version 2.3.22PR1.0 (Casa Software Ltd). All spectra were charge corrected and shifted with respect to the CTi 1s peak at 281.8 eV. [13,28,29] SAXS and WAXS profile of the fibers were measured (in air) using the SAXS/WAXS beamline at the Australian Synchrotron (Melbourne, Australia). Analysis was carried out using a Pilatus3-2 M detector, [54] at 18.2 keV beam energy (wavelength 0.6812 Å) and sample-detector distance of ≈738 mm. The data was processed using the in-house data managing software scatterBrain ver. 2.82. From the 2D scattering patterns, the Herman's orientation factor (f) which describes the degree of orientation of the (002) peaks of the nanomaterials relative to the longitudinal fiber axis was calculated using expression: where the mean-square cosine is derived from the scattered intensity in the azimuthal direction I(ø) through integration over the azimuthal angle ø (extracted from the sector plots of the SAXS/WAXS pattern using scatterBrain ver. 2.82) and is defined as: where azimuthal angle (ø) is the angle between fiber plane and the MXene sheet plane. Therefore, if f = 1 all MXene sheets are parallel to the fiber plane and f = 0 for randomly oriented MXene sheets. The integration of the sector plots was calculated using a ø range of −90° to 90°. The fibers cross-sectional area observed in the SEM images were estimated using ImageJ adapted from a previously reported method ( Figure S5 and Table S1, Supporting Information). [14,28] These crosssections were used for normalizing the electrical conductivity, mechanical properties, and electrochemical performance. A total of 10 spots per sample group were used for estimating the cross-sectional area and the average and standard error of the measurements were reported (Table S1, Supporting Information). The linear resistance (R) of the fibers was calculated using the slope of I-V curve acquired using a multi-meter (Keysight B2901A, UK) with a custom-built four-point probe set-up (2.5 mm probe spacing). The R values were then converted to conductivity (S cm −1 ) using: where the l stands for probe spacing as length and A is the area of the fiber cross-section.
Mechanical properties of the fibers were measured using a universal testing machine (UTM) (Agilent T150 USA) with a 0.5 N load cell. Fiber samples of 2 cm length were mounted onto paper frames with 10 mm windows and tested using a strain rate of 5.0 × 10 −4 s −1 (or 0.005 mm s −1 ).
The electrochemical properties of the fiber electrodes were measured in a three-electrode setup attached to a potentiostat (BioLogic SP-300, France). As described in previous works, the setup uses 1 m H 2 SO 4 as electrolyte, Ag/AgCl (in 3.5 m KCl) as reference electrode, and graphite rod (diameter: 6 mm, length: 2 cm) as the counter electrode. Electrochemical properties of 1-cm long fibers including cyclic voltammetry (CV) and galvanostatic charge-discharge (GCD) curves were recorded at scan rates varying from 2 to 500 mV s −1 (CV) and a current density of 2 A cm −3 (GCD). The volumetric capacitance (C V ) of the wet-spun fibers was calculated from the CV curves according to the Equation (5): where I is the current, v is the scan rate, ∆U is the potential window, and V is the volume of the fiber electrode (accounting for the corrugated fiber surface by using the SEM cross-section area). The capacitance retention of the fiber electrodes were calculated through Equation (6) by using the specific capacitance (C i ) in the final cycle i and the specific capacitance in the first cycle (C 1 ).

Capacitance retention 100%
Lastly, electrochemical impedance spectroscopy (EIS) was measured in an open circuit potential (OCP) using an alternating current (AC) voltage with an amplitude of 10 mV and a frequency range of 10 mHz to 200 kHz.
Agitation Resistance and Stability in Dispersion: Agitation tests were carried out by immersing a 1 cm-long fiber sample (M 100 , M 97.5 , M 95 , and M 90 ) into deionized water followed by exposure to ultrasonication. A digital photograph was taken every 30 min of exposure for a total time of 60 min. Stability of the fiber samples (M 100 and M 95 ) in dispersion was tested by mounting the fiber samples on a glass slide using strips of polyimide tape (Kapton) followed by subsequent exposure in either air, PBS, or PBS at 40 °C. The fibers were then vacuum dried for a span of 2 h prior to each measurement and then re-exposed to air, PBS, and PBS at 40 °C afterward. The electrical resistance of fibers was monitored daily over a span of seven days.
Cytotoxicity: The cytotoxicity of the fibers was evaluated using media extracts from a culture of THP-1 monocytic cells. Additional details on how the cells were cultured are presented in the Supporting Information. Fibers of ≈2 cm in length were immersed in 100 µL of cells (≈9.8 × 10 4 cells per well) in 96 well plates, followed by incubation at 37 °C for 24 h. Negative and positive controls were prepared by incubating RPMI-1640 media only and THP-1 cells. A colorimetric assay, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) (MTT) assay kit (Sigma, CGD-1), was used to measure the cell viability. Briefly, 100 µL of extract/control medium was transferred to a clear well and MTT solution (5 mg mL −1 , 10 µL) was added. Samples were incubated for 1 h at 37 °C, and 5% CO 2 and 100 µL of MTT solvent (acidified isopropanol) was added to each well to dissolve the tetrazolium crystals formed. The absorption/extraction was set to 570 and 690 nm to correct for background absorbance using Varioskan LUX plate reader. All cell culture experiments were conducted in triplicate and standard curves were prepared to calculate cell number. Using the corrected absorbance, cell viability was calculated by using the formula:

Supporting Information
Supporting Information is available from the Wiley Online Library or from the author.