Development of conductive textile fabric using Plackett–Burman optimized green synthesized silver nanoparticles and in situ polymerized polypyrrole

ABSTRACT Electronic textiles (e-textiles) are undergoing rapid technological advancements to attain e-textiles that look and feel like conventional textile fabrics. Research seeks to develop highly functionalized textile-based sensors, actuators, and energy storage devices that integrate seamlessly with current textile technologies. Presently, developments are limited by either low electrical performance, or high cost and complex construction. Additionally, negotiating the balance between high performing e-textiles and environmentally benign production is a challenge. In this report, green synthesized silver nanoparticles (AgNPs) are composited with the conjugated polymer, polypyrrole (Ppy), to create a low-cost conductive textile fabric. A Plackett–Burman design of experiment was used to optimize lime peel extract (LPE) mediated reduction for the synthesis of AgNPs. The results of this optimization process revealed silver nitrate concentration to be a significant factor in both size and UV-vis absorption maxima of the LPE-synthesized AgNPs, and reaction temperature also affecting UV-vis absorption maxima. The resultant optimized AgNPs were consistent in size (40–80 nm) and dispersity (PDI = 0.250). The LPE-synthesized AgNPs are used to form a AgNP-Ppy nanocomposite with a linen textile to produce an e-textile with low electrical resistance (37 Ω). GRAPHICAL ABSTRACT


Introduction
Recent advancements in electronic textiles (e-textiles) have led to the development of high performance, fully integrated wearable textile-based sensors and actuators.Several researchers have used conducting polymers such as polypyrrole (Ppy), perfluorononanesulfonic acid (PFNS), and poly (3,4-ethylenedioxythiophene) polystyrene sulfonate (PEDOT:PSS) to develop e-textiles (1)(2)(3)(4).Another approach is the implementation of nanomaterial coatings.Researchers have used materials such as manganese dioxide nanowires/copper nanowires on cellulose fiber, graphene and PEDOT:PSS dip coated on cotton, nanosoldered carbon nanotubes on non-woven fabrics, and high-temperature thermally annealed graphene film embroidered on to cotton fabric (5)(6)(7)(8).Numerous studies have established the development of nanofunctionalised conductive textile fabrics, but so far, there has been very little attention paid to the use of green synthesized nanomaterials in conductive textile fabrics (9).
Considerable research has established that silver nanoparticles (AgNPs) possess superlative electrical conductivity, and established biocompatibility (10,11).AgNPs have been used in biomedical devices, drug delivery, multimodal cancer therapy, wound dressings, SARS-CoV-2 inhibition, and antimicrobial clothing (10)(11)(12)(13)(14)(15)(16)(17).Research has shown that AgNPs are not toxic to eukaryotic cells but are toxic to prokaryotic cells, hence their success as antimicrobial agents (10,11).Despite this, there remains a lack of consensus on AgNPs' toxicity to health and the environment (10,13).Aboelmaati et al. conducted in vivo research to find that green synthesized AgNPs were more biocompatible than conventionally synthesized AgNPs (13).Conventional AgNP synthesis methods use toxic precursors and high energy costs (10).The results of Aboelmaati et al. demonstrated that the research animals had a higher tolerance to the green synthesized AgNPs as compared to conventionally synthesized AgNPs (13).
Green synthesis is defined as an environmentally conscious method of chemistry intended to utilize ecological solvents (such as water) and eliminate toxic waste products (10,18).Dutta et al. used lime peel extract to reduce silver nitrate (AgNO 3 ) to AgNPs and analyzed their antimicrobial efficacy; their results report AgNPs with an average diameter of 107 nm and a polydispersity index (PDI) of 0.250 via zeta analysis (19).Pugazhenthiran et al. used lime peel extract to synthesize silver quantum dots (QDs) and assessed the efficacy as a biocidal agent to cancel cells (20).Transmission electron microscopy (TEM) was used to determine that the silver QDs were homogenous with an average particle size of 2.6 nm; PDI was not reported (20).Green synthesis methods are commonly limited by issues with polydispersity and repeatability (21).To address this, several studies have optimized the green synthesis process using statistical approaches such as Taguchi method, Box-Behnken design, and Plackett-Burman (22)(23)(24).The Plackett-Burman design in particular is an efficient and economical method to screen large quantities of variables to detect main effects (25).
Ppy is a stable, biocompatible conducting polymer that facilitates straightforward polymerization processes (26).Research by Katouah et al. developed AgNP-Ppy coated fibers by coating cotton in AgNO 3 , ammonium acetate and pyrrole (27).The developed e-textile attained conductivity of 5.23 × 10 S cm −1 (27).A similar process was used by Firoz Babu et al. during which a Ppy silver nanocomposite coated cotton e-textile was developed by impregnating cotton in pyrrole and then adding AgNO 3 as a polymerization initiator (28).The conductivity of this development reached 4.82 × 10 −3 S cm −1 (28).Lv et al. developed a Ppy and micro silver coating cotton spandex by adding pyrrole, sodium 5-sulfosalicylate and FeCl 3 to the fabric for in situ polymerization (29).
Electrodeposition was used to prepare silver 'flowers' on the Ppy surface which delivered pressure sensing functionality with a sensitivity of 17.41 kPa −1 up to 900 kPa (29).
Herein, an e-textile is presented that was functionalized with green synthesized AgNPs and Ppy, and constructed using facile methods that are concurrent with current textile industry dyeing methods (30)(31)(32).A plant based textile, linen, was selected as the base textile due to its low energy requirements in production compared to petrochemical-based textiles (33,34).The green synthesis method was optimized using a Plackett-Burman design of experiment (DOE).The electrical resistance, wash durability, and physicochemical characteristics of the resulting e-textile were analyzed.

Lime peel extract (LPE) preparation
Fresh lime peels were cut into small pieces (approximate size 3 × 5 mm) and added to DS water (1:10 w/v) in a conical flask and heated to 100°C for 20 min.The extract was then filtered and decanted into a Duran flask to be stored in the fridge at 4-5°C until required (Figure 1).

Plackett-Burman experimental design
Optimization of the AgNP synthesis was carried out using a Plackett-Burman method.Initial experimental results from the LPE-mediated synthesis demonstrated the quintessential color change which signified the successful formation of AgNPs (37,38).This change occurred between the 3 and 5-h mark.Data reported by Tarannum et al. (10) alongside these experimental results were used to define the process parameters and levels for the Plackett-Burman design (Table 1).Minitab software was used to create and analyze a 5 factor, 2 level design with 20 runs and 1 center point (Supplementary Information, Table S1) (10,39).The responses obtained were subject to analysis of variance (ANOVA) and regression analysis.

E-textile development
The full experimental design is shown in Figure 2. The process consisted of AgNP synthesis, and AgNP-Ppy application on linen textile.

AgNP synthesis
AgNPs were synthesized by developing the methods first reported by Pugazhenthiran et al. and Dutta et al. (19,20,40).Each experiment used differing conditions (i.e.volume, AgNO 3 concentration, synthesis time, pH, and LPE volume) based on the requirements of the Plackett-Burman design shown in Table 1.LPE (10 ml) was added dropwise to silver nitrate solution (30 ml or 60 ml).The pH was measured and adjusted using a sodium carbonate solution (0.1 M or 1.0 M was used depending on the requirements of each run) to achieve the desired pH.Concentration and pH were set as per the requirement for Plackett-Burman run.The solution was then centrifuged (5 min, 3,000 RPM) to remove any biomass and the precipitate was discarded.Supernatant was centrifuged in DS water (60 min, 6000 RPM), the precipitate was rinsed and centrifuged again in DS water or methanol (20 min, 6000 RPM).Precipitate was either removed and redispersed in DS or removed and then dried in the oven at 70°C.

Ppy AgNP coating on linen
The Ppy-AgNP coated-linen (PAL) textiles were prepared as follows (Figure 3 and Table 2).Three linen fabrics (10.5 cm length, 3 cm width, approx.1.50 g) were dip coated in both pyrrole and AgNPs and polymerized in situ using the following parameters (Table 2).PAL-Py3 used a smaller 0.5 g linen sample.

Characterization
AgNP formation was confirmed via UV-vis spectroscopy and the absorption peak was recorded for each AgNP DOE run (UV/Vis/NIR Spectrophotometer, Jasco V-700 series).Where there was no obvious peak, the nearest lambda (λ) max was recorded for the experiment to contribute to the Plackett-Burman analysis.Oftentimes, the nearest λ = 350 nm.A sample of AgNP was quantified to determine the green synthesis yield.This was assessed by drying the AgNP solution in an oven and weighing the sample.The linen sample thickness was measured using a digital thickness gauge (Testex, TF121).AgNP size and PDI were analyzed using a Malvern Zetasizer (Nano ZS).Elemental analysis was conducted using Energy Dispersive X-ray Fluorescence Spectrometer (EDX-XRF) (Shimadzu EDX-7000 XRF).Compounds within the e-textile were analyzed via Fourier Transform Infrared spectrometry (FT-IR, Jasco FT/IR-4600 series) using a KBr pellet attachment.Scanning electron microscopy/energy dispersive x-ray spectrometry (SEM/ EDS) analysis was conducted using a Jeol JSM-6610LV Scanning Electron Microscope coupled with an Oxford Instruments INCA X-max80 Energy Dispersive X-Ray Spectrometer.Each sample was coated with 30 nm of gold (Au) using a Quorum Q150RS coating unit prior to SEM/EDS analysis.

Sample preparation for FTIR analysis
Prior to FTIR analysis, a small cutting from the textile sample was ground in a pestle and mortar with potassium bromide (KBr) before being pressed into a pellet using a hydraulic press (Specac, Mini-Pellet Press).

Electrical resistance analysis
Electrical resistance was measured using a LCR bridge (Rohde and Schwarz HM8118) at point distances of 1, 2, 5 cm and the full length of the sample (10.3-11.5 cm).Sheet resistance was calculated using the following equation: Where R is the electrical resistance, W is the width of the sample and D is the distance between the two electrodes (41, 42).

Electrical resistance testing under bending
The response of the e-textile sample under bending conditions was assessed qualitatively by attaching copper plating to each end of the sample and connecting to a microcontroller via crocodile clips.

Wash durability analysis
Wash testing was carried out using a washing fastness tester (Testex, TF418) following the BS EN ISO 105-  C06:2010, A1S standard, without steel balls.Following the test, the samples were rinsed and dried for 10 min in the oven at 70°C.

UV-vis analysis
UV-vis is the principal characterization technique to confirm AgNP synthesis.The unique surface plasmon resonance (SPR) resulting from the collective electron oscillation resonating with light found in AgNPs causes characteristic absorption maxima in the UV-vis spectra at higher wavelengths than equivalently sized non-plasmonic particles (10,43).A broad absorption peak is indicative of a polydisperse collection of AgNPs and/or agglomeration, whilst a narrow peak indicates a monodisperse collection (43,44).A shoulder in the spectra below the lambda max peak in AgNP synthesis is characteristic of interband transitions caused by the excitation of inner valence electrons (45).
The nature of the Plackett-Burman design is to use the extremes of "reasonable" parameters within each condition, and therefore it is not designed to find the optimal solution through the experimental runs.Therefore, as anticipated, many of the parameters forming the Plackett-Burman design did not form AgNPs, and this can be seen by the range of spectra in Figure S1 (Supplementary Information).However, there is a collection of spectra which peak at around 420 nm, which is well within the expected range for AgNPs of between 391 and 453 nm (46).The optimized AgNP consistently produced an absorption maxima at around 420 nm.Step 1 A linen sample (3 × 3 cm, 0.5 g) was placed in pyrrole solution (1 M, 53 ml) and stirred at room temperature (RT).
Step 2 After 30 min, AgNP (18 ml) solution was added.After 41 h, the linen was removed from the solution, rinsed with distilled water, and left to try on filter paper in the dark.
The reaction was stirred slowly for 91.5 h and covered with parafilm and aluminum foil.
Step 5 After 91.5 h, the linen was removed from the solution, rinsed with distilled water, and left to dry on filter paper in the dark.Step 6 Coating weight: 0.687 g/46% Coating weight: 0.234 g/23% Coating weight: 1.016 g/51%  Size, shape, and capping agents all influence the SPR and optical properties of the AgNPs and thus UV-vis spectra (43).

Zeta size & zeta potential
The AgNPs were further characterized through zeta size analysis to confirm the size of the nanoparticles (Table 3 and Figure 4).The optimized AgNPs consistently produced AgNPs between 40 and 80 nm with a PDI of approximately 0.250 indicating a homogenous solution of AgNPs.When a methanol rinse and drying stage was added to the filtration process to aid separation, an increase in the average size of AgNPs was seen.

Plackett-Burman optimization
The Plackett-Burman method is a cost-effective method to determine the main effects of the green synthesis method.Whilst a full factorial DOE could have provided the most comprehensive information, fully accounting for the effects of interrelating factors, the design  required 243 (3 5 ) treatment conditions, which was cost and time prohibitive.In Plackett-Burman analyzes, two-factor interaction is confounded and therefore does not distinguish the effect of interactions between variables from the main effects.
The results produced by the Plackett-Burman method to develop an optimized AgNP solution with a UV-vis peak at 400 nm, PDI of 0.2 and a hydrodynamic diameter of 43 nm are detailed in Table 4.The experimental results are close to the solution estimate, GREEN CHEMISTRY LETTERS AND REVIEWS but the zeta size exceeds the estimate.This is likely due to variations in the pH of the reaction medium.Under certain parameters, the pH changed during the reduction process at a rate that was faster than attainable through the manual pH adjustment.Research into the kinetic and mechanistic processes has proposed that there are two simultaneous mechanisms that form AgNPs: pH-dependent autocatalytic reduction where Ag + is formed on the surface of Ag 2 O; and pseudo-first order kinetic directed reduction of Ag + , which is pH-independent (47).It has been demonstrated that more basic conditions support faster reactions via the formation of Ag 2 O to which AgNPs become adsorbed on to the Ag 2 O surface (47).From here, reduction processes undergo autocatalysis (47).AgNPs synthesized in more acidic conditions tend to be more polydisperse (47).

UV-vis peak absorption (nm)
The Plackett-Burman analysis revealed that there were two significant factors affecting the UV-vis absorption of the LPE-synthesized AgNPs: Temperature, p = .022;and the AgNO 3 concentration, p = .059,as shown in Table S2 (Supplementary Information).

Size (nm)
The size of AgNPs were significantly affected by the AgNO 3 concentration where p = .003(Table S3,

Supplementary Information
).This finding is consistent with previous observations in the literature (37,(48)(49)(50).One hypothesis for this relationship is that increasing AgNO 3 concentration leads to a decrease in the magnitude of zeta potential closer to zero, thus indicating a decreased stability in the AgNP resulting in agglomeration (48).In the case of AgNO 3 which disassociates into Ag + and NO − 3 , the increase in AgNO 3 molar concentration is directly proportional to an increase in ionic strength of the reduced silver solution.An increase in ionic strength causes the electric double layer to compress leading to a decrease in the zeta potential (51).

PDI
This study found no statistically significant effect on the PDI response from the factors under analysis (Table S4, Supplementary Information).This result is contrary to previous findings that demonstrated a positive correlation between polydispersity and AgNO 3 concentration between 0.01 and 0.05 mM,  and above 0.25 mM, and a negative correlation from 0.05 to 0.25 mM (37).This differing result may be due to insufficient levels in the Plackett-Burman design to detect the sensitive response of polydispersity in relation to AgNO 3 concentration.
It is notable that there are several interactions affecting the responses.For example, there is an evident interaction between the pH of the reaction and the AgNO 3 concentration in terms of the polydispersity of the final product (Figure 5).Further analysis would be required to separate these factors.

Ppy-silver nanoparticle in-situ synthesis
In the present work, when FeCl 3 was added as a polymerization initiator to the reaction, the solution immediately changed from the colloidal brown of the AgNP to a clear solution.Following this, it gradually changed to a dark green/gray and through to black (Figure S2, Supplementary Information).When AgNO 3 was added as the polymerization initiator, this colour change was less intense.The polymerization of pyrrole via FeCl 3 or AgNO 3 results in Ppy with NO − 3 or Cl − dopants in the polymer chain, respectively (52).The favored hypothesis for the polymerization of pyrrole via AgNO 3 is where an electronic transfer is undergone between the Ag + ions and the pyrrole monomer.This results in the simultaneous synthesis of Ppy and Ag 0 which will then be deposited on the linen surface (52).The Ppy interacts with cellulose via the hydrogen in the NH group, or via lone electron pairs of the pyrrole nitrogen, and the lone electron pairs of cellulosic hydroxyl groups (52).

FT-IR characterization
The FT-IR spectra confirmed the successful polymerization of pyrrole in situ on linen (Figure 6).The essential OH stretch is visible in all samples at around 3420 cm −1 from the linen fabric, although a blueshift is observed in all coated samples (27).An absorption maxima at around 2900 cm −1 is visible in PAL-10 and PAL-Py3 which is assigned to CH stretching, but in samples PAL-11 and PAL-12, a doublet appears signifying the presence of aldehyde (27).The peak at around 1640 cm −1 reveals the in-plane OH vibration of absorbed water molecules (27,52).The reduction in peak intensity around 1030 cm −1 may indicate a carbonyl interaction between the linen and the AgNPs (53).A peak at 1455 cm −1 is visible in some samples indicating the C = C vibration of pyrrole rings (Figure 6(a, b)) (54).The peak at 1540 cm −1 indicates the C = C vibration of the pyrrole rings (Figure 6(a-d)) (54).The maxima peak at 1430 cm −1 demonstrates aliphatic CH wagging mode (Figure 6(b, d)) (27).A strong peak at 1384 cm −1 can be assigned to the NO stretch of the nitrate counter ion in the AgNO 3 polymerized composites (Figure 6(a,b)), indicating doping of the Ppy structure with nitrate anions (28,32,52).This cannot be found in FeCl 3 synthesized composites as the Ppy doping occurs with chloride ions (52).A peak at around 1315 cm −1 is assigned to C-N stretching vibration and the consistent peak at 1029 cm −1 is due to C-H in plane vibration (Figure 6 (a-d)) (32).The out of plane CH vibration is also visible at a peak around 890 cm −1 (Figure 6(c, d)) (54).

Elemental analysis (EDX-XRF)
The chloride doping of Ppy polymerized by FeCl 3 is confirmed through SEM-EDX analysis.In PAL-11 and the Ppy only coated linen sample (FeCl 3 polymerized), the presence of Cl is evident, whereas in PAL-10 and PAL-12, no Cl is visible, but Ag, N and O are abundant.This is further supported by EDX-XRF analysis (Figure 7) which shows PAL-11 to have a substantial percentage of Cl content but also reveals 1% silver content.It is worth noting that the apparent high proportion of silicon in PAL-py3 is relative to other components; the inorganic percentage is considerably smaller than in the other samples at just 0.46% (Figure 8).The overall Si content within the samples ranges from 0.1 to 0.5%.This is likely to be the remnants from the lubrication of the linen fabric during the weaving  process since the plain linen sample also contained 0.3% Si.

Morphology (SEM-EDX)
The morphologies of PAL nanocomposites are shown in Figures 9-12.The classic globular morphology of Ppy can be clearly seen in the SEM micrographs (Figures 9  and 11) corroborating with the morphological findings of previous studies (55,56).Figures 10 and 12 show a morphology which may reveal the presence of silver flakes embedded in the Ppy matrix.Figure 12 showing PAL-12 is particularly interesting where it appears that a film is smoothing over the highly silver-loaded surface.PAL-12 was synthesized by adding pyrrole and then AgNP and then AgNO 3 , whereas PAL-10 was synthesized by adding AgNP and then pyrrole and then AgNO 3 .The SEM-EDS results, in correlation with that of the XRF, indicated that PAL-10 comprises a Ppy film with approximately 0.62% content of silver content at the point of measurement.The flakes in the micrograph (Figure 10) seem to indicate silver plate morphology in PAL-10, whereas PAL-12 showed a Ppy film which coated the 1.88% content of silver (57).

Electrical resistance
The PAL samples exhibited excellent electrical resistance that is competitive with the current state of the art, as demonstrated in Table 5. PAL-11 had the lowest electrical resistance at 9.56 × 10 1 Ω/sq..The electrical resistance of each sample increased significantly as the measurement point distances were increased, with PAL-11 increasing by 732% from 1 cm point distance to 10.5 cm point distance.This change in electrical resistance may be due to a lack of uniformity in areas of the coating which may have led to barriers to both phonon and electron transport (6,58,59).This lack of uniformity may have occurred due to the way that the 10.5 cm samples bent and moved within the beaker during polymerization.The samples were subject to constant agitation but occasionally got stuck in certain positions.Comparing this increase in electrical resistance with other studies is difficult since most do not report the electrical resistance at multiple measurement points.

Electrical resistance under bending
The PAL samples continued to demonstrate low electrical resistance under bending and twisting conditions.
The electrical resistance exhibited change during this analysis which indicated a potential for sensing functionality as shown in Figure 13.Further analysis of these properties is part of ongoing research.

Wash durability analysis
The electrical resistance of four PAL samples was analyzed following the test, rinse and drying, and found a significant increase in electrical resistance (Table 6).This suggests that the current nanocomposite coating is not able to satisfactorily withstand the rigors of domestic and commercial washing.Whilst these results were not encouraging, the findings support evidence from previous observations (55,65).
The doping (Cl − ) of the Ppy affects the charge distribution and delocalization across the polymer, decreasing its electrical resistance (67).The 6.5% reduction of Cl content from PAL-11 reflects the deprotonation at the NH moiety of Ppy under the basic environment of the reference detergent (pH 10.5-11.5),converting the conducting polymer into a non-conducting base (55,(68)(69)(70).The pK a value of Ppy tends to range between 9 and 10, which facilitates its deprotonation in an alkaline environment, increase in the band gap and the observed increase in electrical resistance (Figure 14) (55,65,67,68,71,72).In PAL-10 and PAL-12, a significant decrease in silver content is seen, indicating the partial leaching of AgNPs, potential via the release of Ag + ions when in contact with water (73).This study hypothesized that Ppy would offer a degree of protection to the AgNPs during wash processes due to the hydrophobicity that Ppy imbues.However, the results do not support this.

Conclusions
The purpose of this work was to develop a reliable and consistent method to develop e-textiles whilst minimizing negative impacts on the environment.A green synthesis method was used in the development of AgNPs, and a plant-based textile fabric was used.A Plackett-Burman DOE was used to optimize the LPE synthesis.This found that AgNO 3 concentration and reaction temperature were statistically significant in the lime peel-mediated AgNP synthesis.This research has shown that LPE can effectively be utilized to synthesize homogenous AgNPs tuned to a size of between 40 and 80 nm, with a PDI of 0.250.LPE-synthesized AgNPs were incorporated onto linen fabric as a nanocomposite with Ppy to develop an e-textile with low electrical resistance (9.56 × 10 1 Ω/sq.).Inorganic elemental analysis (excluding the linen fabric and Ppy) revealed that PAL-11 which exhibited the lowest electrical resistance contained 1% silver content and 92% chlorine content (due to the Cl− counterions in Ppy), indicating successfully doped AgNP-Ppy linen e-textile.Under wash testing conditions, the electrical resistance of the e-textile sample increased but remained electrically conductive.The e-textile sample continued to exhibit electrical properties under bending, indicating potential for their use in sensing applications.Overall, these results suggest that AgNP-Ppy based e-textiles are a promising route for the development of highly conductive, environmentally benign e-textiles that can be produced using simple, scalable processes.Further research might usefully explore the physicochemical structure of AgNP-Ppy on linen and also optimize the synthesis of Ppy with AgNP to enhance the percolation network and durability under washing conditions.Additionally, a further study could assess these new e-textiles to study the electrical and mechanical properties, and sensing functionality.

Figure 7 .
Figure 7. Inorganic elemental analysis (EDX-XRF) of PAL samples excluding organic content (linen and Ppy).(AgNO 3 ) and (FeCl 3 ) refers to the polymerization route used in the sample.

Figure 8 .
Figure 8. EDX-XRF analysis of organic vs inorganic elements (AgNO3) and (FeCl3) refers to the polymerization route used in the sample.

Table 2 .
Parameters for polypyrrole silver nanoparticle synthesis on linen.

Table 3 .
Zeta size analysis result detail.
Zeta size analysis of Plackett-Burman Run 6 Z-Average (d.nm)

Table 5 .
Comparison of electrical resistance in e-textile reported in the literature and current study.