A Green Electrically Conductive Textile with Tunable Piezoresistivity and Transiency

A green textile‐based conductor with controllable electrical resistance change with deformation and transiency (i.e., dissolution in water) will be the holy grail in wearable electronics since it can satisfy divergent needs with a single solution and be sustainable simultaneously. Nevertheless, designing such material is challenging since opposite requirements shall be satisfied. To solve such a problem, cotton is functionalized using conductive inks made of graphene or carbon nanofiber, a biodegradable polyvinyl alcohol binder, and environmentally friendly solvents. The electrical resistance shows an anisotropic response to bending depending on the composition of the coating and the stress direction, functioning either as a deformable compliant electrode or a tunable piezoresistor. Indeed, it can withstand thousands of bending cycles with a change in resistance of less than 5% or change its resistance by many orders of magnitude with the same deformation thanks to the combination of cotton twill and different nanofillers. A simple modification in the binder composition adding waterborne polyurethane allows the coating to go from entirely transient in water within minutes to withstanding simulated washing cycles for hours without losing its electrical conductivity. This green versatile conductor may serve opposing needs by altering the material composition and the deformation direction.

The ORCID identification number(s) for the author(s) of this article can be found under https://doi.org/10.1002/adfm.202301542. energy harvesting are all being revolutionized by wearable electronics. [1][2][3][4][5][6][7] The market for this industry was ≈80 billion euros in 2020 and is anticipated to grow to ≈470 billion euros in 2030. [8] At the moment, the most popular wearable technology is the smartwatch, while the smart glass is expected to be the next major innovation. [8] These technologies rely on traditional rigid, bulky, and long-lasting materials for electronics engineered with an ad-hoc wearable design and in contact with the human body. Electrically insulating, conductive, and semiconducting fabrics assembled as electronic textiles (e-textile) are still futuristic. Nevertheless, large-scale produced e-textiles are needed to translate the existing commercial device to a generation in an intimate relationship with the body and enable comprehensive monitoring of body movements or physiological parameters. [7] Integrating common electrical parts like integrated circuits and batteries within textiles is one method for creating an e-textile. [9,10] This approach is simple but is limited by the mechanical properties and washability of the rigid and stiff electronic components. [2,10] The second generation of e-textile is designed to have the textile itself to be the component/device to realize fiber-based electronics or sensors directly on the fabric or through a weaving of functional fabrics. [10] Wearable electrical conductors are required for this last generation of e-textiles to provide signal transmission, electrical connection, and device prototyping.
Knitting, weaving, and sewing metal wires into textiles are popular methods for producing wearables that are electrically conductive. [1,9] Metals, however, lack the wearability and flexibility required in such applications. [10] Functionalizing standard fabrics with conductive materials is another popular approach. [2,3,7,[11][12][13][14][15] With methods like screen printing, dip-, spray-, and blade-coating, solution deposition of inks or pastes is efficient for large-area functionalization of textiles at ambient temperature and pressure. [1,3] Surfactants that can also act as binders can be used to increase the stability of ink dispersion. [3,14] Such surfactants and binders frequently come from unsustainable oil sources. Organic nanocarbons, conductive polymers, or micro-and nano-sized metals like silver, gold, and MXenes are widely used in the production of these inks and

Introduction
The monitoring of physiological signals, how we engage in sports, the fashion industry, human-machine interfaces, and pastes. [3,9,12,[16][17][18] When compared to nanometals, MXenes, and conductive polymers, nanocarbons stand out for their variety of structure (i.e., 1D, 2D, or 3D), low cost, lightweight, and ease of dispersibility in environmentally friendly solvents. [19][20][21][22][23][24] In particular, carbon black, carbon nanofibers (Cs), or graphene nanoplatelets (Gs) ensure tens of tons scale availability and reasonable pricing compared to fullerene, single-wall carbon nanotubes (CnTs), or single-layer graphene. [13,[19][20][21][22]24] Another benefit of 1D and 2D nanocarbons is their high aspect ratio, which ensures an efficient electrical percolation network at low loadings. [3,[19][20][21][22]24] Since signal transmission and electronic conduction depend on the integrity of the conductive paths, wearable interconnects require the stability of the electrical performance of the conductive textile upon deformation and washing. This function is very demanding because wearable devices experience high levels of mechanical stress and water contact while being used and washed. [25,26] E-textiles that use natural textiles as substrates and biodegradable biopolymers as matrices/binders of the conductive fillers are increasingly investigated. [27][28][29][30] Such materials are suitable for green electronics because of their natural abundance, biodegradability, and sustainability. [27][28][29] However, since biopolymers are inherently not made to withstand extreme mechanical deformation and washing cycles, using them is more challenging than using conventional durable materials for electronics. [4,27,31] One of the most relevant examples is the work by Liang X. et al. [32] in which a conductive water-based ink made of CnTs and silk sericin was used to dye textiles. Another recent work in this context is the one from Wang B. et al., [12] in which cotton fabric was functionalized with MXene and carboxymethyl chitosan binder by a layer-by-layer eco-friendly method. In many works in this area, washability is not even mentioned, or it is ensured with encapsulation with a protective petroleum-derived plastic layer. [17] In other works, [33][34][35] graphene oxide was used to functionalize natural fibers or textiles. However, to obtain conductivity, postprocessing with chemical or heat-based reduction were required. These treatments often involve using toxic substances and a large amount of energy and can damage the fabric. [3] Conductive hydrogels made with natural and degradable polymers have been also proposed as green wearable conductors. [36][37][38][39][40][41] Yet, hydrogels have ionic conductivity that cannot sustain a constant electronic current in a circuit. [42][43][44] Furthermore, the ionic conduction features depend on hydration and, as such, stabilizing rubbery coating are used to "freeze" the hydrogel system. [42,43,45] In literature, there is also a growing interest in developing green piezoresistors utilizing degradable materials. [30,41,[46][47][48][49][50][51][52][53][54] In contrast to wearable interconnects, piezoresistors require a linear, large, and repeatable change in resistance over a wide range of strains. [19,24,35] Additionally, there are various requirements inside the piezoresistors category itself. For example, monitoring human movements needs an exact match between deformation and resistance change, whereas structural health monitors require more of an "On/Off" feedback with a massive resistance variation with small deformations. [21] Degradable hydrogels are one method for creating green piezoresistors, which are used to track human movement. [38,48,55] Nevertheless, such technologies have hydrogel-related problems (i.e., ionic conductivity and dependence on hydration). Other approaches are pencil writing [56,57] and carbonization of biomass. [48,58] Both methods still struggle to create sensors that exhibit repeatable behaviors when bent. Functionalizing natural substrates such as cotton or cellulose with green conductive coatings, as described for wearable interconnects but targeting a high gauge factor this time, is promising. [59,60] As before, the drawbacks are washability and harsh postprocessing that can damage the substrate.
Transient electronics, which Rogers J.A. et al. first presented in 2012, is one rapidly developing field in green electronics. [61] When exposed to degrading agents, the most common of which is water, transient devices should disaggregate/degrade partially or completely within a predetermined time range. [61][62][63] Transient electronics are expected to alleviate the burden of the electronic waste (e-waste). [64,65] In fact, using a deteriorating version of a material in some electronic devices can help reduce the production of e-waste. [62,64] However, since textiles must resist numerous washing cycles, transiency in wearable electronics and sensing is challenging. A controllable transiency depending on the application should be developed.
A green wearable conductor tunable and adaptable in terms of change in resistance with deformation and, to some extent, in transiency would be ideal since it could satisfy divergent needs with a single solution. This would bring us closer to on-demand degradable wearable conductors/piezoresistors for electronics. Nevertheless, succeeding in such a task is highly challenging since diametrically opposite needs should be satisfied.
Here, we functionalized cotton using conductive inks of Gs or Cs, a biodegradable binder, and environmentally friendly solvents made mixing water and isopropanol. The electrical resistance of the functionalized cotton showed an anisotropic response to bending depending on the stress direction and the composition of the coating itself. The coatings made with Cs remained stable when bent. Deformations significantly altered the electrical resistance of layers with Gs. With the same bending radius, the cotton weave allowed for tuning of the degree of deformation of the coatings. For instance, when the bending of the coated fabric occurs in direction parallel to the fabric's twills, the carbon nanofibers-based coating can function as a conformable conductor with little (i.e., <5%) variations in the initial resistance. In contrast, the cotton fabric functionalized with the graphene nanoplatelets-based coating is excellent for structural health monitoring when it bends transverse to the fabric's twills since it increases resistance by orders of magnitude. A simple change in the binder composition incorporating a water-resistant polymer allowed the coating to go from completely transient in water within minutes to withstanding simulated washing cycles for hours without losing its electrical conductivity. Such green and sustainable technology may serve opposing needs by altering the material composition and the deformation direction, finding a universal solution to different requests in fields such as robotics, wearables for sports monitoring, and the structural sensing sector.

Results and Discussion
Only environmentally friendly solvents, a biodegradable polymer as a binder, and large-scale available and cheap conductive nanofillers were used in the design of the wearable conductor. Specifically, a mixture of water and isopropanol (H 2 O to IPA ratio of 2:1), polyvinyl alcohol (PVA), and graphene nanoplatelets or carbon nanofibers were selected. Since it is more environmentally friendly and totally dissolves PVA, water is preferable to IPA. [3] The ink formulation was optimized by employing a concentration of 12.5 mg mL −1 of PVA and adding different amounts of nanofillers ranging from 2 to 50 wt.% to the dry biopolymer. IPA facilitates the deposition process on a cotton substrate, based on spray, and increased the stability of the nanofillers' dispersion. [3] The procedure of preparation and deposition of the ink is schematically described in Figure 1a. The dimensions and Raman peaks of the carbon-based materials were characterized previously. [66,67] Briefly, the Gs have an average thickness and lateral size of 6 to 8 nm and ≈ 8 µm, respectively, while the Cs have an average diameter and length of 100 nm and 20 to 200 µm, respectively. [20,67] The PVA was melted and the inks' adhesion to the cotton substrate was increased when the conductive inks were heat pressed at 180 °C after being sprayed on the substrate. The TGA measurements in Figure S1 (Supporting Information) demonstrate that the materials used in this manufacturing process are not being degraded upon hot pressing. The resulting coated textile preserves the flexibility of the uncoated substrate. The chosen S-twill weave structure of the pure cotton was maintained after coating with the inks, as shown in Figure 1b Figure S2, Supporting Information). In the direction parallel to the twill of the weave, the "height" is constant, following the crest of a twill or the bottom of a valley between two parallel twills. Such structure is fundamental for the resulting electrical conduction features and piezoresistivity, as discussed in Figures 2 and 3. The micro morphology highlights a more flat surface of the Gs-based coating compared to the Cs one, in Figure 1d,e, respectively, confirmed The electrical percolation threshold defines the minimum quantity of conductive nanofillers inserted in a polymer matrix to obtain a conductor. [19,43,68] The percolation threshold can be controlled by the nanofiller's aspect ratio, dispersion, and  chemistry. [68] The sheet resistance of the coating is shown in Figure 2a with increasing loading (wt.%) of carbon nanofillers relative to PVA mass. For the graphene-containing layer, the sheet resistance is exceeding 10 9 Ω □ −1 up to 15 wt.% concentration. It declines six orders of magnitude between 15 and 20 wt.%, demonstrating that the percolation threshold is between these loadings. With further rising graphene loadings, the sheet resistance decreases, reaching values in the order of a hundred Ω □ −1 at 50 wt.%. In contrast, the carbon nanofiberbased coatings present a reduced percolation threshold compared to the graphene-based ones. Indeed, the sheet resistance decreases several orders of magnitude between 4 and 8 wt.% of nanofiller. Increasing the amount of nanofibers leads to a further reduction in the sheet resistance, reaching the order of magnitude of 10 Ω □ −1 at 50 wt.% concentration. Considering the thicknesses extracted from the SEM cross sectional images in Figure S2 (Supporting Information), the resistivity of the samples with the highest concentration of nanofillers are 1•10 −4 Ω m and of 4•10 −4 Ω m for the carbon nanofibers and graphene-based samples, respectively. Note that the obtained sheet resistance and resistivity are comparable with or better than previous works based on carbon-based nanofillers coating on cotton, which range between 10 8 to 10 Ω □ −1 and 10 3 to 10 −5 Ω m, respectively. [3] It should be noted that most previous works do not use green solvents and/or binders. [3,69,70] Including conductive fillers into the cotton coating brought additional advantages, such as improved thermal features. As shown in the Figure S3 (Supporting Information), the thermal diffusivity increased by over 3-fold compared to the pure cotton for the highest Gs loading, while the Cs-based coating led to an improvement of less than 2-fold. Smart thermal clothing will be essential in adapting to the fast-changing climate environment. In such a scenario, 2D and 1D green conductive inks can represent an eco-friendly, economically viable solution for increasing the thermal comfort of textiles.
Considering the low sheet resistance of the samples loaded with 50 wt.% of both graphene nanoplatelets and carbon nanofibers, the rest of the characterizations were performed on such samples, unless specified differently. From now on, 50G and 50C labels are used for samples with 50 wt.% of graphene nanoplatelets or carbon nanofibers with respect to the dry weight of the biocomposite coating.
The percolation measurements were made by drawing with silver paint contacts that are not oriented in any specific direction with respect to the twill of the cotton fabric. To investigate if the particular cotton structure described in Figure 1 and Figure S2 (Supporting Information) has some anisotropic effect on the sheet resistance of the fabric, we drew conductive silver tracks in the direction parallel or transverse to the twills' direction, as shown in the schematic of Figure 2b (green or yellow outline, respectively). The G-coated cotton exhibits a strong anisotropy of the electrical properties depending on the direction of the measurements. In particular, in the parallel direction, the sheet resistance is close to 180 Ω □ −1 , while in the transverse one, in the order of 1 kΩ □ −1 . The C-based coating also exhibits an anisotropy of the sheet resistance but less pronounced compared to the G-ones, changing from ≈38 to ≈47 Ω □ −1 . We believe that the anisotropy is due to the diverse path length, resulting from the constant or undulating (crest-valley) path that the electrons must follow in the case of the parallel or transverse direction, respectively. The 50C samples show a lower anisotropy due to the morphology of the nanofibers, which bridge the crest-crest bypassing the passage of the electron through the valley, as shown in the cross sectional SEM images of Figure S2 (Supporting Information).
To verify if and how the cotton texture and the use of Gs or Cs influence the piezoresistivity of the green electrical conductor, we performed repeated bending cycles with the bend direction parallel or transverse to the thread, as schematized at the top of Figure 3. The measurements were taken from samples in the bent position with a radius of curvature of 2.5 mm (labeled as "curved" in the figure), as shown in Figure S4 (Supporting Information), and when the specimens were back to the original position ("flat" label in the figure, no curvature). All the samples show a change from the first bending cycle (See Figure S5, Supporting Information). With the bend direction parallel to the twills, the curved 50G sample exhibits a change in the initial value of resistance R 0 of ≈55% after 20 000 bending cycles, while the percent variation is below 5% in the flat one, as displayed in Figure 3a. In contrast, R 0 of 50C varies below 5% in both curved and flat configurations, as shown in Figure 3b, revealing high stability under deformation. With the bend direction transverse to the textile weave, in the case of the G-based textile, the ratio of R/R 0 increases to 3·10 4 in the curved configuration, while it is still below 5% in the flat one (see Figure 3c). Conversely, in the case of the C-based materials, the R 0 increases of roughly seven times in the curved configuration after 20 000 bending cycles, while the resistance only increases of ≈5% in the flat mode (Figure 3d).
The mechanism behind such piezoresistivity is a mixture between crack propagation and slippage disconnection inside the nanofiller network. [3,71] Tunneling is not plausible at this high concentration since nanofillers directly touch each other. We propose that the tunable piezoresistivity of the functionalized textile under deformation is given by the combination of the following factors: 1) The wavy configuration of the cotton substrate: the valley-hill structure of the substrate can facilitate the creation of significant cracks (or not) if the bend direction is parallel or transverse to the valley-hill arrangement.
2) The mechanical properties of the coatings: the use of Gs or Cs in the coatings composition leads to diverse mechanical features, [66] as shown in the supporting Figure S6 (Supporting Information), which finally may determine the creation of more or less cracks during deformation.
3) The nanofiller aspect ratio: the higher aspect ratio of nanofibers compared to graphene nanoplatelets permits maintaining the percolation of the conductive network more efficiently under deformation.
Such a collection of different behaviors, depending on the bend direction and the nanofiller employed, enables the use of anisotropic textiles in diverse applications. For example, in the case a wearable electrical conductor with near-to constant behavior during bending is needed, the C-coated fabric bended in direction parallel to the twills can be privileged. On the other hand, if structural health monitoring is required, the best www.afm-journal.de www.advancedsciencenews.com 2301542 (6 of 9) © 2023 The Authors. Advanced Functional Materials published by Wiley-VCH GmbH configuration of the wearable conductor would be the G-based fabric used in the transverse mode. Note that in Figure 3a,b, there is a resistance decrease in the flat configuration approaching 20 000 bending cycles. Such a phenomenon may be due to the rearrangement of the nanofillers inside the soft polymer matrix. With increasing bending cycles, the nanofiller gradually forms a better-interconnected network that reduces resistance, partially restoring the initial loss. Similar behaviors were detected and studied with metallic or carbon nanofillers inside soft polymeric networks. [19,72,73] Another desirable property of a wearable conductive material is the preservation of electrical conductivity upon repeated abrasion, since such mechanical stress is expected with daily usage. This feature is normally provided through hard coatings. [20,74] Flexible, lightweight, and wearable fabrics that can withstand repeated abrasion cycles while maintaining their electrical conductivity are still a great challenge to be manufactured. However, they could fulfill unmet needs and enable large-scale use of such textiles. Hence, samples with coatings containing either Gs or Cs were tested against repeated abrasion cycles, as shown in Figure 4a. The 50G samples increase their electrical resistance by one order of magnitude after 200 abrasion cycles, while for the 50C samples the resistance only doubles. At 350 abrasions cycles, the graphene nanoplatelets-based coating of the textile augments its R 0 by 80 times and is not conductive anymore at the 400 th abrasion. Conversely, the carbon nanofibersbased coating of the textile is still conductive after 1000 abrasions, increasing the initial resistance by 28 times at this stage. Comparing these results with similar previous research findings, we can conclude that the conductive textiles developed in this work show a remarkable abrasion resistance. For example, in our previous report, [20] a mixed coating of Gs and Cs and one with only Gs on top of cellulose, after 180 abrasion cycles, showed an order of magnitude increase in the electrical resistance and no conductivity, respectively. In contrast, as discussed previously, both the G-and C-based coatings presented here exhibit better performances. The remarkable abrasion resistance of the coating is due to the hot-pressing procedure that melts the PVA and ensures a significant interaction and an impregnation of the conductive layer inside the fibrous interlace. [69] Furthermore, we believe that the Cs-based coating better preserves the conductivity upon abrasion than the Gs-based coating due to the higher aspect ratio of the carbon nanofibers compared to the graphene nanoplatelets, which permits maintaining percolation easily even with partial removal of the coating with abrasion cycles, as shown in Figure S7 (Supporting Information).
Having the possibility to tune the rate of dissolution of the electrically conductive material with water and, therefore, varying their electrical resistance at a controlled pace depending simply on the material composition is challenging. Still, it could provide a suitable solution for a conductor in diverse application contexts. For example, electrodes for wearables need to withstand multiple washing cycles without any modification in their properties. [3,63,69,[75][76][77][78][79][80] In contrast, some wearable transient electronics need devices that perform a function and then dissolve, partially or fully. [3,63,69,[75][76][77][78][79][80] To tune and control the dissolution rate with water of the conductive layer, we added waterborne polyurethane (PU) to the 50G and 50C coatings to make them more resistant to water, using two formulations, i.e., PVA:PU at 2:1 and 1:1 ratio. PU was chosen because it is known for having good miscibility in water with PVA and to form hydrogen bonding between its carbonyl group and the hydroxyl groups of the PVA. [81] We tested the electrical resistance variation with simulated washing machine cycles. [19,69] The pure PVA samples with either G or C were not conductive after a single washing cycle, removing all the coating from the pure cotton substrate, as shown in Figure S8 (Supporting Information). In contrast, the samples containing PU were all able to withstand 10 washing cycles. In details, the 50C coatings better preserved the conductivity, increasing the R 0 by 2 and 6 times in the case of PVA:PU ratio of 1:1 and 2:1, respectively. For the 50G specimens the R/R 0 after 10 washing cycles was calculated 6 and 20, in the case of PVA:PU ratio of 1:1 and 2:1, respectively. This result indicates that we can tune the solubility of the conductive coating depending on the amount of PU added to the coating. The C-based coatings of the cotton textiles preserve better the electrical conductivity since the higher aspect ratio of the carbon nanofibers compared to the graphene nanoplatelets help sustaining the electrical percolation network. Such performances are similar to previous state of the art, but in our work, both the binders and the solvents used are more environmentally friendly. [3,69,82] Note that in the case of pure PVA used as the binder, after one single washing cycle, the pure cotton substrate could be recycled for further use, coated another time with fresh conductive inks.

Conclusion
Green wearable conductors have been presented based on functionalized cotton fabrics that can function as either a deformable compliant electrode or a tunable piezoresistor depending on the composition and the direction of bending of the fabric. Cotton was sprayed with conductive inks that contained either carbon nanofibers or graphene nanoplatelets and a biopolymer binder. The ink is green since it was made using a water-alcohol mixture as a solvent and the biodegradable PVA as a binder. The substrate's wavy anisotropic structure and the use of various nanocarbons make it possible to adjust the piezoresistivity in accordance with the direction of the bend. For instance, the cotton fabric coated with the graphene nanoplatelets-based coating is excellent for structural health monitoring when it bends transverse to the fabric's twills texture (see Figure 3 for the scheme) since it increases resistance by four orders of magnitude when it bends with a 2.5 mm bending radius. In contrast, when the bending of the functionalized fabric occurs in direction parallel to the fabric's twills, the carbon nanofibers-based coating can function as a wearable conformable conductor with little (i.e., <5%) changes in the initial resistance. By changing the binder composition incorporating water-borne polyurethane into the ink, the transiency of the conductive layer onto the cotton can be modulated. With pure PVA binder, it is possible to have a completely transient coating that disappears after tens of minutes of washing cycles. Changing the ratio of PVA and PU and the nanocarbon utilized, the change in the initial resistance value after ten washing cycles (equivalent to 600 min) can be modulated ranging from two to twenty times. The coating improved the thermal diffusivity of the pure cotton. Such a versatile green wearable conductor could serve as a practical and unique solution for the opposite requirements of wearable electrically conductive materials and be applied as a flexible interconnect, as well as a motion sensor for robots, or as a structural health monitor of motile constructions.

Experimental Section
Materials: A local market supplied the cotton. It had a 1/2 twill weave structure. Graphene nanoplatelets were obtained from XG Sciences (grade M25) and were characterized previously (lateral size 7.7 µm, thickness of 6-8 nm). [67] Graphitized CNFs were acquired from Sigma-Aldrich (grade PR-25-XT-HHT from Pyrograf Products Inc.) and were previously characterized (length ranging from 20 to 200 µm, diameter ≈ 100 nm). [66] Polyvinyl alcohol (molecular weight 500 -5000, 85-89% high purity, hydrolyzed) was purchased from VWR International Ltd. Isopropanol was purchased from Sigma-Aldrich. Air-heat curable water-borne polyurethane was obtained by BASF (grade Joncryl U 4190, aliphatic polyurethane dispersion in water, solids by weight 36.5%, minimum film forming temperature 23 °C). The conductive inks were fabricated in water:IPA solution (2:1 ratio). The conductive ink containing either Gs or Cs was manufactured by dissolving the PVA (12,5 mg mL −1 ) in the solvent mixture and including nanocarbons wt.% relative to the amount of biopolymer, depending on the final load chosen. The maximum amount of nanofiller employed was 6 mg ml −1 for 50C/50G samples. For clarity, to make 12 ml of ink for the 50G sample, 8 ml of water, 4 ml of IPA, 0.15 g of PVA, and 0.075 g of graphene nanoplatelets were employed. Nine milliliter of conductive dispersion was spray-coated on the cotton (35 cm 2 ) after tip sonication (20 kHz, 750 W, 40% amplitude, four times for 60 s) from 15-20 cm distance. In this case (i.e., 50G sample), the amount of coating deposited was ≈ 5 mg cm −2 . Note that the spray nozzle size was ≈ 0.5 mm. Thus, clogging was not happening since the dimension of the nanofiller was much lower than the nozzle size. While spraying, a heat gun was used at ≈100 °C to hasten solvent evaporation. A pressing procedure (3 min, pressure 31.25 bar at 180 °C) was performed using Teflon anti-attachment foil to prevent the sticking of the paper on the platen of the press. Such a pressing procedure ensured the adhesion of the coating to the textile and the best conductivity. The polyurethane was added to the ink in PVA:PU 2:1 and 1:1 ratios to provide the desired washing stability.
Methods: For all measurements mean values were presented, along with error bars, from at least three different samples unless otherwise specified.
Thermal Degradation: Thermal gravimetric analysis was performed on samples of a mass of ≈7 mg with a TA Instruments Q5500 TGA. Under a nitrogen environment, samples were heated to 800 °C at a rate of 10 °C min −1 .
Morphological Characterization: The optical microscope images were acquired using a VHX-7000 digital microscope.
A Zeiss Evo50 microscope working at an acceleration voltage of 10 kV was used to capture SEM images of the morphology and cross section of the specimens (1 cm 2 ). The samples were fractured after being frozen with liquid nitrogen for cross sectional SEM examination.
Electrical Characterization: A source-meter from Keithley (model 2450) in the four-probe configuration was used to measure the electrical characteristics (resistance, R) of the cotton-based conductors. Five mm wide contacts (W) distant five mm (L) were made on the specimens by coating them with silver conductive paste (RS pro, item number 186-3600). As illustrated in the schematic of Figure 2b, conductive silver tracks were drawn in the direction parallel or transverse to the twills direction for the measure based on the orientation of the textile weave (green or yellow outline, respectively). The resistivity was calculated following the formula where R, W, and L were defined above, and t is the thickness. Repeated bending cycles with the bend direction parallel or transverse to the twill were performed as schematized at the top of Figure 3. Samples (2 × 1 cm 2 ) undergo repeatable and uniform bending cycles while suspended between two supports at a 2.5 mm bending radius. One support oscillated horizontally, causing the sample to transition between a flat and a curved configuration (see Figure S3, Supporting Information). A pneumatic cylinder (Festo Model ADN-20-50-A-P-A) which was controlled by an electronically switched solenoid valve (Festo Model VUVG Metric M5 5/2), caused the sliding support to move back and forth. During bending cycles, change in the initial value of resistance in both the flat and curved configuration was measured. Four-point probe surface resistance measurements (four 20 µm thin copper wires were attached to the edges of the sample in a rectangular arrangement) were taken at regular intervals and normalized to the initial value of resistance. Surface resistance values were obtained by sweeping a DC current (Keithley Model 6221) between two contacts and measuring the resultant voltage across the other two contacts (Keithley Models 2182A). During each sweep, the current flow was always along the direction of motion of the oscillating horizontal support. Mechanical Characterization: An Instron 3365 (pull rate of 10 mm min −1 ) was employed for measuring the stress-strain features of the cotton-based conductors.
Resistance to Abrasion: On a Taber Industries 5155 Abraser with CS10 abrading wheels and a 250 g total mass load per wheel, abrasion testing was done. Between each abrasion interval, the wheels were refaced with S-11 disks for 50 cycles. The testing speed was 72 rpm. The samples used in this test were circular and had a 12 cm diameter. The resistance was measured across the abrasion region and normalized to the initial resistance value.
Washing Machine Cycles: The conductive textiles were washed in water (volume of ≈500 mL, dimension of the sample of (7.5 × 5) cm 2 ). Ten one-hour washing cycles were carried out. By using a magnetic stirrer and maintaining the water's temperature at 40 °C throughout the cycles, water movement was kept up. A detergent (Cussons Carex Complete, ≈4 mL) was added for each cycle. [19] The sheet resistance was measured with the setup described above before and after each cycle.
Thermal Diffusivity Measurements: The IR thermogram setup was utilized to determine the effective thermal diffusivity, following the method reported in the literature. [20,22] The setups utilized a broadband pulsed laser (1 Hz) as a heat source to generate periodic heat waves in the material and a high-resolution infrared camera as a detector (FLIR T660).

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