Sage Journals: Discover world-class research

Abstract

This research aimed to verify embroidery parameters for manufacturing high-performance dry transcutaneous electrical neural stimulation (TENS) electrodes for future applications in soft-good end products. The embroidery parameters were verified by measuring and calculating surface resistance and signal-to-noise values for manufactured electrodes. Parameters explored in this study included different conductive threads, stitch patterns, stitch densities, and fabric grain-to-stitch orientations, using a Melco^® Amaya Bravo Single Head 16 Needle Embroidery Machine. The chosen embroidery parameters were guided by Goncu Berk’s work in “Design of a wearable pain management system with embroidered TENS electrodes.” An additional aim of the study was to measure the performance of developed e-broidery TENS electrode samples after laundering stresses, following AATCC Test Method 61. The surface resistivity and signal-to-noise values of e-broidered electrodes were measured pre- and post-laundering. The surface resistivity values were again measured and compared to the pre-laundering data. Findings support that core-spun conductive thread stitched parallel to the bias grain is recommended for longevity and reduced impact of laundering. This study of embroidered TENS electrode performance provided crucial insight into better understanding life span and viability as a future “smart” medical wearable device.

Keywords

CAD Embroidery Launderability Pain Management TENS Electrodes Wearable Devices

Introduction

Transcutaneous electrical neural stimulation (TENS) electrodes are part of a device that emits low-voltage currents to electrically stimulate nerves and block pain sensations. Consequently, TENS has been used as a treatment for various acute and chronic pains. The application of TENS therapy stems from Shealy et al.’s¹ research on neuro-modulation techniques and Melzack and Wall’s² Gate Control Theory. Gate Control Theory suggests that the spinal cord contains neurological “gates” that specifically allow pain signals traveling via small nerve fibers to pass while signals sent by large nerve fibers are blocked. Current commercially available TENS devices are approximately a hand-held size (e.g. mobile phone size range); consisting of a control/power unit, and electrodes hooked to the unit through loose cables. The low-voltage currents are delivered to the skin through hydrogel Ag/AgCl electrodes that stick to the skin. These electrodes depend on their ability to retain moisture in order to function. Currently, TENS hydrogel electrodes lose their moisture after a few applications, inhibiting their intended function and lacking reusability. TENS embroidered electrodes are wearable electronic devices. When such wearable technology is fully integrated into clothing, products are referred to as “smart clothing” or “smart garments,” or “textronics”—a “dynamically developing segment of textile products which combines elements of electronics and textiles.”³ Euler et al.⁴ conducted a review of textile-based electrodes developed for eletrostimulation and reported a variety of production techniques with comparison electrode parameters, one of which labeled e-embroidery (or machine stitching with conductive yarns) as a “Level 3” electrode—the highest level of conductivity option for integration into the textile substrate (not including seamless integration using knitting or weaving methods). This high level of integration along with flexibility of placement and application deems it a preferred option for manufacturing. Manufacturing a product that uses embroidery electrodes on a non-conductive stretch textile surface with a transferable circuit board/power supply has the potential to increase the comfort and mobility of a TENS device for users. Embroidered electrodes can be optimized through computer-aided (CAD) embroidery parameters—such as stitch pattern, stitch density, and stitch orientation.⁵ As suggested by Shuvo et al.,⁶ evaluating the performance of TENS embroidered electrodes after laundering stresses is necessary. Therefore, the goal of this study was to further verify and develop embroidery as a reasonable and reliable means of manufacturing high-performance dry TENS electrodes for soft-good product application while also assessing performance post laundering.

A literature review shows increasing developments for electrodes and sensors using a variety of manufacturing approaches including ink and 3D printing,⁷ textile fabrication,^8–10 soft-circuitry,¹¹ and embroidery techniques^12,13 within the emerging field of e-textiles, particularly for medical use.¹⁴ Innovation in textile-based electronics, commonly referred to as “textronics,” aims to advance the field by developing soft, wearable sensing technologies designed for physiological monitoring, thereby addressing growing demands in healthcare, sports, and human performance applications. Shuvo et al.⁶ conducted an in-depth review of the materials and fabrication methodologies used for state-of-the-art textronic sensors per textile form factors (fiber, yarn, fabric, and apparel) used in the sensor system and suggested future research to focus on mechanical performance of yarn-based electrodes post washing due to findings associated with angular displacement and shrinkage (in additional to other suggestions that align with manufacturing and FDA regulations and pathways). In addition to angular displacement and shrinkage after the mechanical stresses associated with the laundering process is functional conductivity performance reliability. This laundering challenge of smart textronics is due to intimate contact with the human body, which naturally becomes soiled, and must be cleaned/laundered. Therefore, this study focused on mechanical performance of conductive e-broidery TENS electrodes post-laundering.

Post et al.¹⁵ introduced the term “e-broidery”—to describe the process of stitching in a distinct pattern using conductive thread and machine embroidery equipment—as an e-textile fabrication process. Conductive embroidery allows for the control and integration of conductive threads with possibly different electrical properties in one step, making it more appealing than knitting and weaving in terms of manufacturing.¹⁶ Conductive embroidery also allows for better reproducibility of the manufacturing process and rapid prototyping when in tandem with the stitch creation CAD software,¹⁷ allowing this study to investigate more stitch parameters and embroidery techniques. Eike et al.,⁵ plus others, have identified the need to specify via documentation the specific parameters and details for manufacturing conductive embroidery for replication of work to advance research in this field.

Recent work from Akgün et al.¹⁸ on square-shaped (5 cm× 5 cm) e-broidered TENS therapy knee pad design using X-silver yarns and X-Static^® yarns, including 10-cycle laundering simulation, indicated an electrical resistance difference connected to the length ofe-broidered transmission lines between electrodes. Additionally, this research supported e-broidery with silver based on lower electrical resistance values. Erdem et al.¹⁹ fabricated square-shaped (5 cm× 5 cm) sewn and e-broidered electrodes and compared resistivity values to those of self-adhesive electrodes. This research found that silver based textile electrodes have lower resistance values than the carbon version, particularly when embroidered in an “overlapping pattern.” Lastly, research from Skrzetuska et al.³ compared three different textile substrates for comfort (surface mass and air permeability) and performance (friction, stretching, and washing) properties in conjunction with (4 cm × 8 cm) film-printed and e-broidered electrodes (stacking channel shaped) with a metal latch compatible with TENS simulation devices, for potential application in soft products for pain management. However, the e-broidery electrode shape and parameters for e-broidery fabrication lack details that support shape rationale and replicability for the proposed electrode design.^20–25

Electrical conductivity is essential for functional textronics. There are different ways that researchers have gone about measuring conductive function for e-textiles; however, the most commonly used term to evaluate e-textiles and communicate electrical parameters is resistivity, the measure for how strongly a material resists an electric current. For conductive textiles, test methods for measuring electrical surface resistivity are provided by AATCC TM 76. Electrical resistivity can only be defined by the test methods; there is no independent test for determining a true value. Surface resistivity indicates the effectiveness of the textile in carrying electrical currents and function within specific frequencies. When evaluating e-textile conductivity, a low surface resistivity reading indicates a textile that readily allows an electrical current to transmit across the surface of the fabric effectively at a frequency deemed appropriate for its end use. For TENS electrodes specifically, changes in voltage can cause pain to the wearer of TENS fabrics, which is why it is important that no resistance fluctuations are identified. Electrical resistivity is a material property of a substance whose numerical value is equal to the ratio of the voltage to the current density. Consequently, electrical resistivity is used in this study to evaluate the percentage change in the TENS electrodes potential after various exposure conditions.

There are many environmental conditions that may degrade or compromise electronic devices, including the presence of dust and particulate matter, air pollution, high temperature and humidity levels.²⁶ When worn on the body, additional and inevitable soiling occurs by hair, dandruff, tears, earwax, mucus, saliva, sweat, dead skin, and other intimate bodily secretions, much of which are transferred to clothing and textiles. Body soil may negatively impact smart garment functionality (by increasing surface resistivity) but soiling also needs to be removed for hygienic reasons. This is especially true when e-textiles are included in personal protective equipment or used in medical and/or fitness applications. Thus, any future textronics that come in direct contact with the skin must be washable on demand. Consumers generally launder garments through a wet- or dry-cleaning process. Wet cleaning occurs by hand or mechanical washing machines (either in the home or in a laundromat) with mechanical machine laundering being the preferred and most utilized method. Laundry washing machines are harsh and potentially damaging environments for e-textiles. Mechanical stress (abrasion, bending, torsion), thermal stress (temperature), water stress, and chemical influence (detergents and additives) have been identified as key damaging actions of laundering.^22,23,25,27 Much of the current research in the field of e-textiles mainly focuses on the ability to produce e-textiles, with a growing amount of work focused on the ability to maintain e-textiles, which increases the potential end products’ functional reliability and longevity. This study serves to not only verify the ability to produce and replicate the resistivity performance of TENS electrodes as outlined in the study by Goncu Berk²⁸ (one of the first studies to focus on e-broidery to produce TENS electrodes), but in addition, explore the impact of laundering on their performance to evaluatee-broidered TENS longevity.

The purpose of this project was to replicate and verify e-broidery parameters for manufacturing high-performance dry transcutaneous electrical neural stimulation (TENS) electrodes on soft-good end products and to test the performance of TENS electrodes after laundering stresses. The gathered data was evaluated using a paired t-test when comparing the same sample before versus after laundering, and an independent t-test with unequal variances when comparing parameters across different unique samples. The null hypothesis was that there would be no significant difference between parameters, and significance would be observed when the p-value was at least less than 0.05. With this main purpose of research in mind, an additional goal was to propose embroidery parameters and methods for future continuations of this line of research. Findings of this research aims to continue building the body of knowledge for e-broidery techniques, soft textile-based electrodes, and launderability outcomes for future smart/textronic wearable garment or product potential.

Materials and Methods

TENS Electrode Design

To evaluate the embroidery parameters for TENS electrodes, the different embroidery threads, stitches, stitch densities, and grain orientations were evaluated. Using a Melco^® Amaya Bravo Single Head 16 Needle Embroidery Machine, a total of 48 textile electrodes with a diameter of 30 mm were created using two different conductive silver threads, satin and fill stitch at three different density settings, and two grain orientations: parallel to the grain and parallel to the bias. The circular shape was chosen for the replicability of Goncu Berk’s²⁸ study, “Design of a wearable pain management system with embroidered TENS electrodes.” Determining a rationale for rectangular versus circular electrode shape was not included as a design parameter in this study.

Threads

Two types of conductive silver thread were used for this study: Shieldex 235/34 (Silver 1) and LessEMF (Silver 2). These conductive embroidery threads were used for three main reasons. First, these conductive embroidery threads show superior performance in terms of continuous manufacturability and in terms of end-product use. Second, these threads were chosen for the replicability of previous studies by the research team and the target project for replication conducted by Goncu Berk.²⁸ Third, these threads were manufactured in two distinct ways. The Silver 1 thread was coated with a conductive silver coating, and Silver 2 was core-spun with one thin silver thread or wire. Silver 1 is specified to have a linear electrical resistivity of <0.3 Ω/cm, while Silver 2 is specified to have a linear electrical resistivity of <100 Ω/cm. Note: the LessEMF company no longer carries the silver-plated nylon thread for purchase.

Stitch Type

Fill stitches are intended to cover an entire area so that the fabric cannot be seen through the embroidered area. On the other hand, satin stitches are like a zig-zag stitch with a very small stitch length so that it looks like a continuous line. Satin stitches often span the entire width of the area, unlike fill stitches. In general use, satin stitches are employed over smaller areas for a cleaner look while fill stitches are used over wider spans. If satin stitch is used over too wide of a span, it can make the surface more raised, become easily snagged, or cause the threads to droop.

Density

The density parameters used in this study ranked highest to lowest included: 50%, 80%, and 100%. 80% and 100% were included for comparison to Goncu Berk’s²⁸ study, which included a high stitch density and low stitch density with only stitch length included. Note: When setting up the stitching parameters, these different stitch densities were adjusted within the Melco^® embroidery software.

Fabric Grain Orientation

Unlike the comparison study by Goncu Berk,²⁸ this study also explored fabric grain’s orientation’s effect on TENS electrode performance. Grain orientation in fabric refers to the direction of the threads within a textile, which significantly influences its appearance, drape, and durability. In woven textiles, the grain is defined by the interplay of two sets of threads—warp and weft/filling—which intersect at right angles. This creates a stable structure where the grain is most prominent along the length of the warp, leading to greater strength and less elongation in the warp direction/orientation. When a woven fabric is cut on the bias, a 45-degree angle to both the warp and weft yarns, it exhibits increased stretch and a fluid drape, making it ideal for shaping and contouring along the body. Conversely, knitted textiles are formed through interlocking loops of yarn, resulting in a more flexible material in which the yarns form wales (vertical orientation similar to warp yarns) and courses (horizontal orientation similar to weft/filling yarns). In knitted fabrics, the grain is less rigid, as the loops allow for multidirectional stretch, with the orientation of the yarns often affecting the fabric’s overall stretch/elongation and recovery properties. While knitted fabrics have no strict directional grain, cutting on the bias still affects their stretch properties and drape, although to a lesser extent than in woven fabrics. For this study, electrodes embroidered parallel to the lengthwise grain will be referred to as “on the grain,” and electrodes embroidered parallel to the bias grain will be referred to as “on the bias.” Grain orientation/direction of a textile substrate is important to consider as textile/yarn elongation and recovery are different depending on tension from with the grain or on the bias of a textile. This orientation for e-broidery is important to measure for future application of the electrode when applied (stitched) for wearable textile product end purposes. Table 1 outlines the factors studied in this research.

Table 1.

Factorial design of 48 total samples.

Silver 1												Silver 2
Satin						Fill						Satin						Fill
On the grain			On the bias			On the grain			On the bias			On the grain			On the bias			On the grain			On the bias
100%	80%	50%	100%	80%	50%	100%	80%	50%	100%	80%	50%	100%	80%	50%	100%	80%	50%	100%	80%	50%	100%	80%	50%
2	2	2	2	2	2	2	2	2	2	2	2	2	2	2	2	2	2	2	2	2	2	2	2

Non-conductive Textile Substrate

The TENS electrodes were embroidered onto a 70% Nylon 30% Spandex tricot. This fabric was chosen to expand upon its usability in end products and build upon the findings of Goncu Berk,²⁸ who used a “100% polyester knit fabric” (p. 40). A 100% polyester performance cut-away stabilizer was also used to ensure that the electrodes were not under mechanical tension while undergoing the fabrication process or when worn. The stabilizer also aided in limiting distortion and puckering during the embroidering fabrication process on the stretch spacer fabric, per suggestions from Eike et al.⁵ Moreover, there’s the added advantage of creating a more raised surface when using a stabilizer that betters electrode–skin impedance, crucial to TENS electrode function within the end product.

Specimen Preparation

Each specimen was originally designed for embroidery onto a 10″ × 10″ square of fabric in order to be easily hooped for the industrial embroidery machine. The specimens were later cut down to 160 mm × 160 mm to maintain AATCC Test Method 210, which specifies at least 65 mm from the conductive test samples to the cut edge. See Figure 1 for specimen dimensions.

Figure 1.

Specimen dimensions.

Each unique combination of thread, stitch type, density, and grain orientation, resulting in 24 unique specimens, received two replications for a total of 48 samples. Please see Table 2. Parameters of the select images include Silver 1 and 2 threads, satin and fill stitch types, 100% stitch density, and images of both front and back views. Table 2 outlines an example of 100% stitch density of the two threads used in in this study along with stitch type (front and back views).

Table 2.

Photos of select e-broidered conductive electrode specimens.

Conductive thread	Embroidery stitch	Front view	Back view
Silver 1	Satin
Silver 1	Fill
Silver 2	Satin
Silver 2	Fill

All images reflect 100% stitch density; please contact the corresponding author for additional images if desired.

Procedures

Electrical Resistance Measurement

To accommodate the wash procedure, surface resistivity data was gathered following a combination of AATCC Test Method 76 and Test Method 210. Surface resistivity refers to the resistance encountered by electrical current as it travels along the surface of the conductive area. Functional checks of surface resistivity were performed on 160 mm × 160 mm unlaundered specimens and 160 mm × 160 mm specimens after being exposed to laundering conditions. Surface resistivity was measured using an Agilent/HP 34401A Multimeter. Resistance of the leads—which measured to be ∼0.045 Ω—was removed during analysis calculations. Laboratory conditions were recorded to be 74.6–80.1°F and 14–20% relative humidity throughout the resistivity data collection process. Resistivity measurements were performed from 16 different point locations per sample, see Figure 2(a). Averages and standard deviations of resistivity measurements per e-textile specimen are reported (please review Table 4 for specimen resistivity values). The resistance values and their distributions for each electrode are compared according to thread type, stitch type, density, and grain orientation using Excel. The results of this study were also compared to those reported by Goncu Berk.²⁸

Figure 2.

(a) Resistance measurement points on TENS electrode surface and (b) line schematic of SNR data collection set-up.

Signal-to-noise Measurement

The signal-to-noise ratio (SNR) is a critical metric used to quantify the strength of a desired signal relative to the background noise. A higher SNR indicates a clearer signal with minimal interference from noise, which is essential for accurate data interpretation. To measure SNF, the power of the signal is first determined by isolating the relevant signal components, while the noise is quantified by evaluating the variance or standard deviation of the background fluctuations. The SNF is used to assess the quality and reliability of the measurements, with a higher SNR generally indicating more accurate and interpretable results. A signal generator was used to create a 10 V, 60 Hz, square waveform signal, similar to a widely used TENS signal for pain management. A digital oscilloscope collected the signals from the specimens at six different point locations, see Figure 2(b). Laboratory conditions were recorded to be 76–77.6°F and 16–23% relative humidity throughout the SNR data collection process. The SNR was calculated by finding the mean value of the signal and dividing it by the standard deviations, then converting it to decibels. (Please review the Table 5 for specific SNR values for specimen measurements.) Data collected from each electrode was calculated and compared Berk’s²⁸ results. The SNR was recorded to better evaluate how the TENS electrodes would operate within the end product device.

When fabricating electrodes from conductive thread via the embroidery technique, the SNR becomes particularly important in ensuring that the electrical signal transmitted through the textile substrate is distinguishable from any noise or interference. In the case ofe-broidered electrodes, variations in surface resistivity can affect the clarity of the signal, and by extension, the SNR. Measuring both the SNR and surface resistivity aids in understanding the properties of the e-broidered electrodes which enables optimization of the fabrication process to achieve high SNR and reliable performance in practical applications. See Figure 2 for data collection schematics.

Laundering Procedure: Rotawash

The Rotawash accelerated laundering system was utilized in this study to carry out the washing simulation of five home launderings in one wash cycle. The Rotawash simulation of five home launderings is appropriate as consumers will generally reject apparel products that fail after being used/washed only five or six times.²⁹ Small textile specimens (rather than whole garments) were laundered in individual rotating canisters, which allows for the control of water in conjunction with textile and additive.

Home, commercial, and most laboratory wash systems use tap water, which varies in mineral content, chlorine content, water hardness, and other factors that affect laundered textiles. In order to reduce the number of confounding factors related to wash water quality, which may have a substantial effect on end conductivity results, distilled water was used to carry out the laundering tests via the Rotawash accelerated laundering system as opposed to traditional washing machines that require a tap water supply line.

The accelerated laundering procedure follows AATCC Test Method 61, Test Condition 1B. Test Condition 1B was selected for its gentler mechanical action, lower water temperature, and higher concentration of detergent. This test condition is most similar to a delicate home washing machine cycle. The selected 31°C temperature is considered a “cold” wash temperature. Table 3 outlines the laundering procedures conducted in this study.

Table 3.

Test conditions of 1B AATCC TM61.

Test no.	Temperature		Total liquorvolume (mL)	Percentage powder detergent of total volume	No. of steelballs	No. ofrubber balls	Time (min)
Test no.	°C (±2°)	°F (±4°)	Total liquorvolume (mL)	Percentage powder detergent of total volume	No. of steelballs	No. ofrubber balls	Time (min)
1B	31	88	150	0.37	0	10	20

Each specimen was laundered separately in a steel canister with a 150 mL solution of distilled water and AATCC test standard powdered detergent without optical brighteners, plus 10 rubber balls. E-textiles tumbled in Rotawash canisters for 20 min at 31°C (88°F). They were then rinsed with distilled water and air-dried flat on lab tables. The changes in resistivity and SNR were then assessed after laundering.

Results and Discussion

Surface Resistivity

Surface resistivity values were recorded from 16 different point locations for two replications of 24 unique sample parameters for a total of 48 specimens and 383 unique data points. The average resistance values for each parameter—stitch type, stitch density, and conductive thread type—for the silver-based textile electrodes specimens and the conventional hydrogel Ag/AgCl electrode can be reviewed in Table 4. To reiterate, the gathered data was evaluated using a paired t-test when comparing the same sample before versus after laundering and an independent t-test with unequal variances when comparing parameters across different unique samples. The null hypothesis was that there would be no significant difference between parameters, and significance was observed when the p-value was less than 0.05.

Table 4.

Resistance values before and after laundering.

Conductivethread	Embroiderystitch	Thread length	Average resistancebefore (Ω)	Averageresistanceafter (Ω)	Standard deviationbefore	Standard deviation after
Silver 1	Fill	High density	14.03	1.05	25.66	1.09
		Medium density	1.35	1.11	0.97	0.98
		Low density	1.4	1.15	0.91	0.92
	Satin	High density	61.93	104.19	82.70	163.21
		Medium density	16.83	89.51	26.10	132.33
		Low density	18.08	84.20	30.36	144.70
Silver 2	Fill	High density	164.11	178.12	223.97	232.64
		Medium density	167.35	305.49	215.24	583.56
		Low density	265.48	249.85	229.85	205.33
	Satin	High density	113.46	162.04	136.05	274.74
		Medium density	132.75	175.42	119.32	151.03
		Low density	213.87	307.87	173.79	306.66
Hydrogel Ag/AgCl			649.25	—	328.17	—

Compared to manufacturer-specified linear resistivity, the accuracy of the surface resistivity measurements varies across the different configurations of conductive threads, embroidery stitches, and densities. Silver 1 satin stitches, particularly at high density, exhibitsubstantially less accuracy compared to manufacturer-predicted resistivity values, indicating potential imprecision in the resistance measurements for these configurations. Similarly, among Silver 2 satin stitches, medium- and low-density configurations showed greater accuracy compared to their high-density counterparts. Both threads showed less accuracy with a high-density satin stitch. The combination of high thread density and the satin stitch might create a more intricate and closely packed structure, making it challenging to manufacture and achieve consistent electrical conductivity. This could result in higher variability in resistance measurements and reduced accuracy.

Furthermore, Silver 1 fill stitches at medium and low densities demonstrate relatively low standard deviations, indicating a higher degree of precision and consistency in the resistance measurements for these configurations. On the other hand, for Silver 2, the fill stitches at all densities present considerable standard deviations, implying a notable degree of variability in the resistance measurements. The observed difference in precision between Silver 1 and Silver 2 when using a fill stitch can be attributed to the distinct manufacturing processes and properties of the two types of conductive threads. Silver 1 is coated with a conductive silver coating, providing a more uniform and consistent coverage of conductive material throughout the thread. The core-spun construction of Silver 2 might introduce variations in the thickness or distribution of the conductive core within the thread. In a fill stitch, where a higher density of thread is used, these variations could result in less predictable electrical conductivity and, consequently, reduced accuracy. See Table 4 and Figure 3 for surface resistivity before and after laundering results. Further analysis and consideration of factors influencing measurement accuracy are warranted to provide a comprehensive understanding of the observed variations.

Figure 3.

Comparison of resistivity before and after laundering.

Threads

Based on the results (review Figure 3), it was determined that Silver 1’s surface resistivity was significantly less/lower than that of Silver 2 (p < 0.001). This was not surprising to the researchers as the resistance reported by the conductive thread suppliers yielded a substantially lower resistivity value (greater electrical conductivity) for Silver 1 compared to Silver 2. The conductive coating on Silver 1’s yarn provided better contact for completing a circuit with the multimeter, compared to Silver 2’s core-spun yarn structure. The other important parameter for TENS electrodes is the homogenous distribution of the resistance values on the surface of the electrodes. A spike in current can cause stinging pain to the user, especially with low resistance values where more voltage is delivered to the skin. Silver 1’s homogeneity was significantly greater than Silver 2’s before and after laundering (before: p < 0.001, after: p < 0.01). However, Silver 1’s surface resistivity significantly increased (p = 0.03–0.05) and homogeneity significantly decreased (p = 0.03–0.05) after laundering. It is speculated that Silver 1’s coating is “flaking” or being removed during the laundering process, compromising the sample’s conductance. Future research needs to study wastewater post-laundering as silver nanoparticles can be detrimental to the environment. Silver has antibacterial properties that although more hygienic for the wearer are harmful to good bacteria within an ecosystem.

On the other hand, Silver 2’s surface resistivity (p = 0.0504) and homogeneity (p = 0.051) were not significantly affected by laundering. Additionally, Silver 2’s resistivity values were closer to that of a conventional hydrogel Ag/AgCl electrode. This comparability between Silver 2 and conventional electrodes was also observed by Goncu Berk.²⁸ This suggests that in terms of reliability and longevity, Silver 2 may be the better alternative.

Stitch Type

When comparing satin versus fill stitch, it was found that satin stitch’s surface resistivity was significantly increased (p < 0.001) and homogeneity was significantly decreased (p < 0.001) by laundering, while the fill stitch was not significantly affected (p = 0.4), as seen in Figure 3. This may be because, over the wider spans on the electrode, the satin stitch is more raised than the fill stitch; possibly exposing it to more stresses when undergoing the laundering process. The wider spans also make manufacturing of the electrodes more difficult. Difficulties and discontinuities during the manufacturing process not only negatively affect the conductance of the sample, but also make it challenging to replicate the end product in an efficient/productive manner. For longevity of the TENS electrode and replicability of the device, it can be recommended to focus on fill stitch for future studies.

Density

The density of the stitches did not show a significant relationship to any other variable. Goncu Berk²⁸ also observed that “the relationship between stitch densities and resistance values do not show a consistent result between Silver 1 and Silver 2 threads and different stitch patterns,” although when looking at the threads individually, Goncu Berk²⁸ observed that lower density had lower resistances. The results from this study showed that the threads, Silver 1 and Silver 2, individually had lower resistance values with higher densities but not significantly so. Based on the analysis and comparison of the results between the two studies, it could be concluded that density does not play a significant factor in TENS electrode performance.

Fabric Grain Orientation

Electrodes embroidered “with the grain” of the fabric experienced a significant increase in surface resistivity (p < 0.01) and a significant decrease in homogeneity (p < 0.01) after laundering. Conversely, “with the bias” specimens were not significantly affected by the laundering process (resistivity p = 0.14, homogeneity p = 0.19). This suggests that the longevity of the smart garment may be extended if conductive embroidery is stitched parallel to the bias grain; e-broidered stitches align with the bias orientation of the yarns. This may be because the yarns in the bias orientation are already under slight tension in spacer fabrics, reducing the mechanical tension on the electrodes. Also, because the bias grain is already under tension in spacer fabric, stitching parallel to that grain (bias orientation) could cause the electrode to be less affected by dimensional changes in the fabric after initial laundering compared to parallel to the lengthwise grain (with the grain; with the knit wales).

Signal-to-noise Ratio

TENS electrodes must be able to transmit specific signals with specific characteristics in order to be effective. Therefore, the SNR must also be evaluated in addition to resistivity in order to make reasonable conclusions in regard to the electrode’s operability within the end product. A typical TENS device functions with an asymmetrical biphasic square waveform, 2–150 Hz pulse frequency, 50–250 µs pulse width, and 0–80 mA current.²⁸ The signal quality of manufactured embroidered electrodes was comparatively analyzed by calculating the SNR values for each electrode. The average SNR for each parameter of embroidery stitch type, stitch density, and conductive thread type for this study’s textile electrodes and the conventional hydrogel Ag/AgCl electrode can be reviewed in Table 5.

Table 5.

Signal-to-noise ratios for textile TENS electrodes and conventional Ag/AgCl electrode.

Embroidery stitch	Thread density	Average SNR before (dB)	Average SNR after (dB)
Fill	High stitch density	17.50	18.27
	Med stitch density	17.38	18.29
	Low stitch density	17.50	17.95
Satin	High stitch density	17.41	18.27
	Med stitch density	17.47	18.15
	Low stitch density	17.32	18.23
Fill	High stitch density	17.53	18.05
	Med stitch density	17.69	18.08
	Low stitch density	17.53	18.05
Satin	High stitch density	17.47	17.88
	Med stitch density	17.57	18.00
	Low stitch density	17.55	17.99
Hydrogel Ag/AgCl electrode		7.42

Interestingly, after laundering, all specimens showed a significant increase in SNR (p < 0.001). Electrodes embroidered against the grain of the fabric also experienced a consistently greater increase in SNR than those parallel to the grain (p = 0.03–0.05). Selection of the optimal embroidered parameters for TENS electrodes requires a thorough understanding of both resistance and SNR values since resistance values do not correspond to SNR values.

In Goncu Berk’s²⁸ study, it was observed that Silver 2 had higher SNR values than Silver 1. This was also observed in this study’s result before laundering. However, Silver 1 experienced a consistently greater increase in SNR than Silver 2 after undergoing the laundering process (p < 0.001). After laundering, Silver’s 1 SNR was relatively similar to Silver 2 as seen in Table 5 and Figure 4. The cause of this is unclear without further research on the surface chemistry occurring on the electrode during laundering.

Figure 4.

Comparison of signal-to-noise ratio before and after laundering.

Although there was an increase in environmental humidity when recording the SNR after laundering, most studies report that an increase in humidity would degrade the signal or decrease the SNR, not increase it. It is possible that the removal of the lubricant used to ease the manufacturing process from the surface of the sample or the residue left on the surface of the sample from the detergent could have impacted the SNR. However, further research is needed to conclude these speculated reasoning’s for the SNR readings.

Conclusion

TENS devices create low-voltage electrical currents that disrupt pain signals and consequently decrease pain sensations, without additional side effects when compared to pharmaceuticals. However, the current design of TENS devices is not ergonomic—reducing the mobility of users with loose, dangling wires—and relies on disposable hydrogel electrodes. A proposed solution is to create electrodes that deliver TENS signals using conductive embroidery thread on a non-conductive textile surface after optimization of embroidery parameters and verifying their reliability or longevity.

In the comparison study, Goncu Berk²⁸ chose Silver 1 satin stitch with low stitch density as the optimal option in terms of TENS electrode performance. Our study did verify many of the same results that Goncu Berk²⁸ observed; however, taking the reliability and longevity of the garment into consideration, Silver 2 fill stitch with 100% stitch density is recommended. The impact of laundering must be considered when developing an end product garment, and Silver 2’s core-spun properties show promise as more reliable longevity than coated threads. Figure 5 outlines a resistivity comparison of the study data.

Figure 5.

Study comparison of resistivity results: Berk versus Depping.

While Goncu Berk’s²⁸ work laid a crucial foundation for understanding e-broidery techniques in the development of TENS electrodes, our study advances the field by exploring a broader range of embroidery parameters and introducing a comprehensive evaluation of electrode performance post-laundering stresses. The identification of Silver 2 fill stitch with 100% stitch density as the recommended option for improved reliability and longevity represents a noteworthy departure from Goncu Berk’s²⁸ findings. The incorporation of investigating fabric grain-to-stitch orientation further distinguishes our approach, offering potential benefits for both electrode performance and ease of manufacturing. By emphasizing these distinctions, our research not only validates and refines Goncu Berk’s²⁸ initial findings but also provides a deeper understanding of the nuanced factors influencing the viability of embroidered TENS electrodes as durable and efficient components of medical wearable devices.

There are many factors that come into play when considering production techniques, such as manufacturing speed, contact with skin, flexibility with electrode design, overall costs, and environmental sustainability,⁴ but our study, comparing outcomes to other researchers’ work of textronic electrodes, provided insight into longevity, in terms of user care (laundering) and performance of the e-broidered electrode. Furthermore, from this study, it was determined that future tests will focus on core-spun threads using a fill stitch for better longevity and easier manufacturing of embroidered TENS electrodes. Density will not be considered in future studies as a research variable affecting electrode performance. However, fabric grain-to-stitch orientation is a significant factor affecting conductive embroidery longevity and should be considered in future investigations.

Although further testing will be necessary in regard to TENS electrodes’ effectiveness at minimizing pain, sensor displacement during dynamic motion, etc., this evaluation of embroidered TENS electrodes’ performance provides crucial progress toward defining their life span and viability as a medical wearable device end product. Additionally, the impedance of embroidered electrodes, a critical parameter for TENS applications,^3,30 will be measured in future research along with resistivity and SNR (per their significance for communication and sensing applications) to further substantiate findings.

In conclusion, the authors invite future collaboration opportunities to explore alternative fabrication parameters and statistical analysis methods for examining the relationships among variables. Future work will involve a comprehensive, frequency-dependent analysis of the impedance characteristics of e-textile electrodes, with the aim of investigating additional components and behaviors relevant to their electrical circuitry. Interested parties are encouraged to contact the corresponding author to discuss potential collaboration on forthcoming projects.

Footnotes

Acknowledgements

Iowa State University Department of Mechanical Engineering Instrumentation Lab is thanked for oscilloscope, multimeter, and signal generator rental. Iowa State University Department of AESHM Quality Assurance Labs is thanked for use of their Rotawash accelerated laundering system. Undergraduate researcher Shane Zenk is thanked for assisting in the literature review.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The research was partly funded by a Stewart Research Award and a University Honors Program Grant through the Iowa State University Foundation.

ORCID iD

Rachel Eike

Data availability statement

The datasets generated for this study can be found in the Iowa State University digital repository associated with Rachel J. Eike (). Datasets are supplemental resources connected to this publication.

References

Shealy

Mortimer

Reswick

JB.

Electrical inhibition of pain by stimulation of the dorsal columns. Anesth Analg 1967; 46: 489–491.

Melzack

Wall

PD.

Pain mechanisms: a new theory. Pain Forum 1996; 5: 3–11.

Skrzetuska

Michalak

Krucińska

Design and analysis of electrodes for electrostimulation (TENS) using the technique of film printing and embroidery in textiles. Sensors 2021; 21(14): 4789.

Euler

Guo

Persson

NK.

A review of textile-based electrodes developed for electrostimulation. Text Res J 2022; 92: 1300–1320.

Eike

Depping

Cliett

Technical considerations and specifications for conductive machine embroidery. Trends Text Eng Fash Technol 2020; 6: 687–696.

Shuvo

Shah

Dagdeviren

. Electronic textile sensors for decoding vital body signals: state-of-the-art review on characterizations and recommendations. Adv Intell Syst 2022; 4: 2100223.

Hou

, et al. Direct ink writing of materials for electronics-related applications: a mini review. Front Mater 2021; 8: 1–8.

Volpes

Valenti

Zafar

, et al. Feasibility of conductive embroidered threads for I2C sensors in microcontroller-based wearable electronics. Flex Print Electron 2023; 8: 015016.

Murali

Muni

BNM

Varma

, et al. Development of wearable antennas with different cotton textiles. Int J Eng Res Appl 2014; 4: 8–14.

10.

Alshabouna

Lee

Barandun

, et al. PEDOT:PSS-modified cotton conductive thread for mass manufacturing of textile-based electrical wearable sensors by computerized embroidery. Mater Today 2022; 59: 56–67.

11.

Gaubert

Boddaert

Djenizian

, et al. Textile electronic circuits from laser-patterned conductive fabric. Adv Eng Mater 2023; 25: 2201548.

12.

Gil

Fernández-García

Tornero

. Embroidery manufacturing techniques for textile dipole antenna applied to wireless body area network. Text Res J 2018; 89(8): 1573–1581.

13.

Wang

Zhang

Volakis

. Textile antennas for wearable radio frequency applications. Text Light Ind Sci Technol 2013; 2: 105–112.

14.

Eike

Bakhshian

. Review of fashion, technology and health: future directions for the apparel industry. Trends Text Eng Fash Technol 2018; 4: 504–507.

15.

Post

Orth

Russo

, et al. E-broidery: design and fabrication of textile-based computing. IBM Syst J 2000; 39: 840–860.

16.

Stoppa

Chiolerio

. Wearable electronics and smart textiles: a critical review. Sensors 2014; 14: 11957–11992.

17.

Mecnika

Hoerr

Krievins

, et al. Technical embroidery for smart textiles: review. Mater Sci Text Cloth Technol 2015; 9: 56.

18.

Akgün

Şenol

Yeşilpinar

. Design of a TENS knee pad with integrated textile electrodes. Ind Textila 2021; 72: 268–273.

19.

Erdem

Yeşilpinar

Şenol

, et al. Design of TENS electrodes using different production techniques. Tekstilec 2016; 59: 132–136.

20.

Uz Zaman

Tao

Cochrane

, et al. Launderability of conductive polymer yarns used for connections of e-textile modules: mechanical stresses. Fibers Polym 2019; 20: 2355–2366.

21.

Uz Zaman

Tao

Cochrane

, et al. Market readiness of smart textile structures – reliability and washability. IOP Conf Ser Mater Sci Eng 2018; 459: 012071.

22.

Uz Zaman

Tao

Cochrane

, et al. Understanding the washing damage to textile ECG dry skin electrodes, embroidered and fabric-based; set up of equivalent laboratory tests. Sensors 2020; 20(5): 1272.

23.

Eike

Depping

Cliett

, et al. Initial insights of laundry detergent and additive impact on e-textile surface resistivity. Text Res J 2023; 93: 3109–3127.

24.

Molla

MTI

Compton

Dunne

. Launderability of surface-insulated cut and sew e-textiles. In: Proceedings – International symposium on wearable computers, ISWC, Singapore, 8–12 October 2018, pp.104–111. Association for Computing Machinery.

25.

Lam

NYK

Tan

Toomey

, et al. Washability and abrasion resistance of illuminative knitted e-textiles with POFs and silver-coated conductive yarns. Fash Text 2022; 9: 39.

26.

, et al. Design of intelligent garment with transcutaneous electrical nerve stimulation function based on the intarsia knitting technique. Text Res J 2010; 80: 279–286.

27.

Eike

Cliett

, et al. Durability testing of electronic textile surface resistivity and textile antenna performance. Text Res J 2019; 89: 3708–3721.

28.

Goncu Berk

. Design of a wearable pain management system with embroidered TENS electrodes. Int J Cloth Sci Technol 2018; 30: 38–48.

29.

Helms

. Assessment of garment performance during use in relation to consumer expectations. Florida State University, https://www.proquest.com/dissertations-theses/assessment-garment-performance-during-use/docview/302603505/se-2?accountid=10906 (1971, accessed 9 April 2024).

30.

Logothetis

Fernandez-Garcia

Troynikov

, et al. Embroidered electrodes for bioelectrical impedance analysis: impact of surface area and stitch parameters. Meas Sci Technol 2019; 30: 115103.

Design and Reliability of Transcutaneous Electrical Neural Stimulation (TENS) Embroidered Electrodes

Abstract

Keywords

Introduction

Materials and Methods

TENS Electrode Design

Threads

Stitch Type

Density

Fabric Grain Orientation

Non-conductive Textile Substrate

Specimen Preparation

Procedures

Electrical Resistance Measurement

Signal-to-noise Measurement

Laundering Procedure: Rotawash

Results and Discussion

Surface Resistivity

Threads

Stitch Type

Density

Fabric Grain Orientation

Signal-to-noise Ratio

Conclusion

Footnotes

Acknowledgements

Declaration of conflicting interests

Funding

ORCID iD

Data availability statement

References