Pertinence of Textile-Based Energy Harvesting System for Biomedical Applications

Abstract

In the era of technological advancements in healthcare and medicine, monitoring of health status and treatment conditions has been made convenient by the development of various categories and forms of biomedical sensors. They have been incorporated within watches and mobile phones and can be worn as stand-alone based on user preference. However, the longevity, cost, and sustainable functionality have impeded its adoption within the population. In this review article, we have introduced a concept of bridging the textile industry and biomedical sensors to yield a self-powered biomedical system that operates on textile-based energy harvesters. Textile-based wearable systems have been compared to E-skin-based systems. The energy released by different actions in human motion has been quantified along with insights on its effective utilization in the form of energy harvesters in the subsequent sections. Information on designing such a textile-based system with schematics has been done. This review focuses on the development and connection of textile-based energy harvesters to existing models of biomedical sensors.

This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

1. Introduction

Traditionally, hospitals with state-of-the-art instruments and patient monitoring facilities have taken precedence when an individual needs aid with diagnosis, treatment, operation, or postoperative care. Monitoring of an individual’s vital organs/systems, such as heart rate, body temperature, and ECG, has been done in the presence of a doctor within a hospital or a doctor’s office room. It is also well known that the treatment of certain medical conditions that necessitate intensive care units (ICU) is expensive not only for the patient but also for the hospital because they must invest in the upkeep of medical equipment such as blood pressure cuffs, oxygen saturation meters, thermometers, breathing, and urinal tubes [1, 2]. Even when the patient is shifted from an ICU room to a general care unit, the majority of the equipment is shifted as well for postoperative monitoring. However, various tubes and needles on the patient body, even after surgery for the monitoring of the vitals, cause a great deal of pain and discomfort as by this stage, the effect of anesthesia given during the operation wears off. This prompted the health-care professionals in the medical field to raise the need for an effective and less expensive setting for the patient’s treatment. Consequently, the integration of nanotechnology, medical, and electronics had opened ways to develop biomedical devices with outstanding features of drug delivery, real-time health monitoring of physiological parameters, and even the recording of nerve stimulation [3]. This gave birth to the smart health monitoring systems that possessed sensors for obtaining medical information from the patient connected to them that can sometimes wirelessly transfer real-time details to the monitor of the health-care professional were known as portable biomedical devices, which revolutionized the health-care system. With the boom of material science and miniature technology, biomedical devices could also be implanted or attached to the skin (wristwatches/smart bandages).

Medical treatments, on the other hand, can sometimes last for months, and in other cases, the patient may need to make the device a permanent part of their life. For such long-term use, biomedical devices had to be built with large batteries, which resulted in substantial models that required frequent replacement, as well as increased consumer discomfort. Moreover, this psychologically creates a negative impact on the wearer’s mind reminding them of the particular medical condition, which has clinically proven to impede recovery rates. Additionally, the apparent visibility of such devices implanted or attached to the skin is hard to miss for an onlooker who quickly identifies the wearer to be a patient or sick, which no wearer finds appealing as for most people their treatment tends to be a matter of privacy. These limitations of traditional biomedical devices have undoubtedly raised the need to develop a self-sustaining power system for a biomedical device to counter the need for frequent replacement of batteries.

The progress of fabrication technology and biocompatibility has seen the textile industry contributing to the biomedical field in intensely effective ways. Due to the material advancements made, biosensors can now be integrated into textiles to make wearable clothes for people [4]. This textile-based system processes and transmits the biosystemic data with vital information directly to the health professional, thus solving the psychological impact of wearing visible biomedical devices and increasing the mental as well as emotional wellbeing, therefore overall benefitting the health of the individual. Textile-based biomedical devices also possess the ability to not only harvest but also self-generate energy.

In this review paper, we aim to emphasize the importance of these textile-based energy harvesters. We begin by highlighting the different categories of wearable health devices and elaborate on the unique features of human motion to understand the energy generated from a human body. This will be followed by the various textile-based energy harvesting systems for application in the biomedical field and an overview of self-powering textile-based wearable systems. The relevant information will be summarized in the final section of this review.

2. Different Types of Wearable Biomedical Devices [WBD]

2.1. E-Skin-Based WBD

Electronics that possess an excellent ability to stretch and showcase elasticity/flexibility with a certain extent of self-healing capabilities are referred to as electronic skin. These are popularly used in the field of prosthetics development, artificial intelligence, robotics, and now even in health monitoring systems [5–8]. E-skin-based WHC devices mimic the capabilities of the human skin and try replicating its function towards external environmental stimuli such as heat, pressure, and temperature.

E-skin, also known as artificial electronic skin, comprises sensors and actuators that receive the biosystemic information and convert it into a form of signal that can be processed, communicate the information, and self-powered. The regions of the skin on which these components are put; biocompatibility and ability to stretch in response to the demands of human mobility, as well as comfort and lightweight for the user with a reasonable shelf life, are all factors in the material selection process. Biofluid-proof materials are required in some instances. The overall device performance of such WBD pertains to the type of materials and substrates used, dimensional structures incorporated into the design, signal processing units, overall architecture, and design formed and size of the storage units. The appeal of E-skin-based WBD devices to potential wearers ultimately depends on functionality but is also more influenced by cost and esthetic appearance.

Typical substrates used for making flexible E-skin-based WBD include a wide variety of polymers segregated and chosen based on Young’s modulus (for flexibility) and degradation temperature (transition temperature, $T_g$) in addition to the factors mentioned in the above paragraphs (refer to Table 1). These include poly-dimethyl-siloxane (PDMS), poly-ethylenetereph-thalate (PET), poly-ethylene-napthalate (PEN), poly-ether-sulfone (PES), poly-ether-ether-ketone (PEEK), polycarbonate (PC), and polyimide (PI) [9–15]. In some cases where the transparency of the appearance of the device needs to be altered, liquid crystal polymer and eco-flex are also utilized as substrate materials that are opaque and translucent in nature [16–18]. Figure 1 illustrates the different types of such WBD devices used in the health-care system.

Table 1

Substrates used to develop E-skin-based WBD.

Substrate material	$T_d{}^{0}C$	$\sigma$$W{m}^{-1}{K}^{-1}$	Transparency	Young’s modulus Pa	References
PDMS	150	0.15-0.17	Transparent	0.3-3 M	[9–13]
PET	120	0.15-0.24	90% transparent	2800-3100 M	[9, 14, 15]
PEN	160	0.15	87% transparent	2500 M	[9, 14–16]
PES	190	0.18	89% transparent	2400-2800 M	[10, 14, 15]
PEEK	240	0.25	NA	3600 M	[10, 14]
PC	150	0.19	92% transparent	2400 M	[15]
PI	>300	1.3	30%-60% transparent	2500 M	[10, 14]
Liquid crystal polymer (LCP)	>350	0.3	Opaque	NA	[17]
Ecoflex	NA	NA	Translucent	0.027-0.029 M	[18–20]

[figure(s) omitted; refer to PDF]

A key component in these WBDs is the range of sensors used. Sensors tend to make or break the entire functionality of the device [11, 12]. The different types of sensors used are demonstrated in Figure 2.

[figure(s) omitted; refer to PDF]

The interaction of the biotic and abiotic components, as well as a robust interface between them, is the key focus for E-skin-based or any skin-based implants [19, 20]. From a manufacturing standpoint, Ray et al. [21] emphasized the necessity of biointegrated devices for health monitoring to be user-specific. There must be excellent compatibility of the implant-/skin-based devices with the texture of the skin, size of the body part, and lifestyle of the wearer. One of many drawbacks of such an approach is the utilization of gel-based adhesive sensors in the manufacturing of skin-based biomedical devices that cause skin irritation. Moreover, large-scale manufacturing of such a technology cannot dwell into the user-specific details; thus, the end product might be of comfort to some but discomfort to others. Apart from this, the dynamic nature of the human biology system that is in continuous contact with rapidly evolving environments poses a significant challenge to the performance of these devices. Moreover, they do not provide continuous health data (real time), making it hard for the health professional to determine the disease onset, which is attributed to the limited power from batteries requiring replacements. This naturally creates the need to develop a counter mechanism that can record and transmit continuous biosystemic data without causing discomfort and involving replacements of batteries. The interconnection between textile-based WBD (parent system) with textile-based energy harvesters as power supply unit is the proposed solution in our work.

2.2. Textile-Based WBD

Textile-based wearable biomedical devices (WBD) had been the invention that was born out of necessity on the battlefield. The concept of an “intelligent garment” was developed since people’s lives depend on immediate medical attention, which is not always possible in a war zone. Primary objectives were to constantly monitor the vital organs of the soldier/wearer with additional features of military benefit including enemy weaponry knowledge by measuring the penetration of bullets on the garment. A group of doctors residing on the military base camp was sent the information directly transmitted from the garment, thus enabling constant monitoring. Smart textiles were born as a result of this concept gaining traction in the fields of material science, electronics, and medicine.

As clothes are worn at all times by people of all age groups, making these clothes, the source of medical information has gained monumental appeal from the biomedical professionals. Unlike noninvasive (electronic skin) and invasive (implants) counterparts that result in psychological stigmatization and discomfort in terms of physical wearability and technical drawbacks as highlighted in the above section, the textile-based health monitoring system provides the wearer with mental, psychological, and overall wellbeing. This is mainly achieved due to the easy incorporation of energy harvesters into the framework design of smart textiles.

From the customer/patient point of view, the functioning of smart textile is as simple as wearing a new shirt. The functionality of such a patient-oriented biomedical invention is summarized in Figure 3.

[figure(s) omitted; refer to PDF]

The sensory unit constitutes arrays of wearable sensors integrated within or onto the textile because the textile is made up of flexible fiber. This allows for the gathering of biomechanical data to be done without interruption while taking into account the natural or movement-induced deformation that occurs in humans. Based on the type of stimuli, sensors could be piezoelectric, capacitive, resistive, and triboelectric, which will be explained later in this review.

Research and development of low-dimensional electronic systems have made it possible to integrate various components onto various substrates and, more importantly, textile products. Simultaneous research into the deposition of coatings onto fibers to synthesize conductive materials/polymers has backed up the former. Further research has now enabled these conductive materials to function as actuators and transducers, which are crucial components of the processing unit shown in Figure 3.

We have taken this one step further and categorized smart textiles into three broad categories, namely, active, passive, and very active smart textiles. Passive smart textiles are believed to be the primitive of the group as these only possess the capacity to make changes to their properties based on stimuli. Optical fibers are popular inventions that come under this category.

Active materials, on the other hand, are the next generation of materials capable of making decisions about how to react to external stimuli. However, the convergence of the textile industry and biomedical field took place on the development grounds of third-generation smart textiles. These not only could sense and make decisions but also configure themselves based on the stimuli, thus opening opportunities to monitor health via apparel and clothing, known as very active smart textiles.

Furthermore, the interpretation and response unit is nearly identical to that of an E-skin-based WBD, with the exception of the level of monitoring, which is not continuous in an E-skin-based WBD due to battery and energy requirements.

Modern society is showing signs of becoming more health conscious. The demand to have a constant update on health and having control over it has pushed even the biggest telecom companies, IT industry, and brands to incorporate health applications within mobile phones and gadgets to make them more appealing. However, these are not built to provide disease onset information, postoperative care, or well-rounded health information. Many medical professionals find data from such devices unreliable due to the vulnerability of glitches.

Thus, the goal of this review is to emphasize the necessity of using the human body as a source of energy and storing it in order to power the WBD, which can only be truly effective with a textile-based approach. The following section will comprehend this approach better by highlighting the enormous amount of energy that can be harnessed from various parts of the human body.

3. Underutilized Marvel: The Human Motion

The calculations of energy given out by the different body parts have been provided by Riemer and Shapiro [22] in their comprehensive work. Our human body per day releases energy equivalent to 20 kilograms of AA batteries with 2500 mAh each, approximately around 800-850 in number. Thermal and movement centered activities are a common outlet for such energy expenses. We can eradicate the adversities associated with modern technology that lack the convenience of long-haul usage attributed to exhaustible power sources/batteries by harvesting this unexploited energy source to power electronics.

The energy to perform human motion is gained from biomolecules such as carbohydrates, proteins, and fats. Energy converted from them by our body for utilization (specific energy) is 35 to 100 times more than the batteries used in the present day, according to Rome, L. C [23]. However, more than 70% of the energy consumed is released towards the atmosphere, thus bringing the mechanical efficiency of the body to less than 30%, validated by Vélez Día and Moreno Gutiérrez [24] in their study. Another reason why the marvel of human motion needs to be explored is to harness all the wasted energy.

According to studies, when a person moves freely, such as during running, energy is generated in various parts of the body. The data in Table 2 depicts this energy in terms of power.

Table 2

Energy expenditure by various parts of the body.

Body movement	Energy released (W)
Elbow	0.78 W
Finger motion	6.9-19 mW
Shoulders	1.34 W
Breathing band	0.83 W
Hip	7.2 W
Knee	33.5 W
Ankle	18.9 W
Heel strike	<10 W
Blood pressure	<1 W

Ultimately, these energies are born and released by the muscles involved in the respective motion. Therefore, to effectively harvest the energy, it is vital to consider the muscular phases—positive muscle work and negative muscle work. These phases occur during a single act of motion.

Any work done by muscles to generate motion is considered the positive phase of work of the muscle and the termination of the motion is the negative phase, in which energy is absorbed to perform this task. Eminent works of Riemer and Shapiro [25] and Alexander [26] validated that the contraction of muscle fibers due to the application of torque at the muscle joint in the same direction of the angular velocity results in motion constituting the positive phase. However, an energy harvester present at this phase to extract the energy developed here would increase the minimum energy required to perform the task, also known as metabolic cost. The body would have to exert more energy to perform the same task which would lead to a detrimental effect on health; therefore, knowing which energy to harvest is crucial. On the other hand, the negative phase involves the opposite of the former. Therefore, an energy harvester placed at this point would hamper no natural motion. Additionally, more energy harvesting can be performed in the regions where the human body gives off energy to the environment.

The heat generated from the human system, according to Carnes, if quantified, amounts to around 100 W, which is released towards the surroundings is underutilized. Thermo-electrical generators can only make use of sensible heat and not latent heat. The temperature difference is the main driving force for sensible heat, and occasions such as sweat evaporation harbor in latent heat. Consider the movement of the arms, the first motion is related to the movement of the elbow (between the forearm and upper arm), and the second motion is related to the upper arm and humorous bone involving the changes in the movement of the shoulder. It was recorded for an average male subject, and the percentage of negative work was between 35-40% for the elbows and 60-63% for the shoulders. An average of 2.2 W of power was generated just by casual walking in these regions, quantified by Winters [27] and his research team.

The most common act of carrying a heavy load on the shoulder may be a bag among children or military troupe or lifting heavy bags of groceries since the power generated as negative work is going to increase for both arm and shoulder. This concept is also called the center of mass motion. In efforts to simplify, it can be best understood as quantifying the energy from a relative motion between the bag and the person carrying it. Research has shown that for each kilogram, 1 W of energy could be available for usage. Thus, the region of shoulders could have energy harvesters, and hands could have special gloves possessing them. This energy could power up WBD that serves monitoring purposes for the upper body in addition to the energy dissipated as heat.

Heel strike, for instance, while walking, occurs each time when the heel strikes the ground. Fair to say, the higher the distance between the heel and the ground higher the energy release. This means that jumping, jogging, and other activities can provide more energy release than walking at a casual pace. Shorten [28], a researcher engaged in quantifying energy loss during heel strike, relied on a previous study [25] that showed energy loss occurs when walking towards the sole. More than 10 J of energy loss occurred based on his calculations when his heel struck the ground. This energy, if harvested, could provide power to the biosensors in WBD. The amount of energy harvested depends on the area of contact from where the loss takes place. In this case of heel strike, socks and shoes are the only two things that provide continuous and maximum contact. Socks and soles of shoes developed with energy harvesters could be materialized through the merging of the textile industry and electronics. Not only can it harvest, but when connected to WBD can provide sustainable livelihood for the long haul as it could probably serve as an inexhaustible source of power to the biomonitoring devices, thus enlightening the importance of developing textile-based energy harvesters. As mentioned in earlier sections, textile products and apparel are the most effective since they are the only components with the most considerable contact with the skin and human body. As a result, establishing a link between textile-based energy harvesters and wearable biomedical devices is the most natural strategy for addressing contemporary difficulties and the key to revolutionizing the biomedical-electronics industry.

4. World of Smart Textile

4.1. Birth and Development of Fabrics

The textile industry has come far from producing fabrics, polymers, and apparel merely for esthetic needs. In the age of personalized health care, personal attire has become the primary medium to deliver medical services. Textiles have been well known to outperform traditional electronics in terms of stretchability and comfort. Modernization has enabled the textile industry to enter an entirely new genre of application by utilizing various techniques to impart functional qualities to their polymers.

A textile-based monitoring system, as explained earlier, requires sensing and transducing functions; these capabilities can be built into the system either by coating it on the textile and using it as the substrate or by using a polymer that itself has transducing capabilities, thus earning itself the name of smart textile. It is essential to understand the hierarchy of processes involved in fabric development. A piece of fabric is an amalgamation of individual fibers (yarn) interconnected in various manners such as knitting, weaving, net formation, tufted, and nonwoven. Additionally, the evolution of composite fabrics enables the fabrication of multi-functional fabrics, formed when two or more types of fabric are interconnected. The following Figure 4 showcases the various forms of fabric development for a clearer understanding [29].

[figure(s) omitted; refer to PDF]

The coating of functional materials (sensing and transducing) can occur during the lower dimensional stage of the hierarchy and, in the final, thus being more flexible than traditional electronics. However, one must keep in mind that each fabric structure has its own properties and will serve a specific application better than others. Therefore, the choice must be based on the common ground where the application meets the properties offered by the polymer.

The geometrical properties of a fiber are its area of cross section and length; the chemical properties of a fiber are determined by its reactivity/inertness to alkaline and acidic environments; and the physical properties of a fiber are determined by the crystallography of the polymer used to make it. Tracton’s [30] and Luo and Van Ooij’s [31] elaborative works provide insightful light on this aspect. The final performance of the coated polymer will depend on the interfacial reaction (between the surface of coating and the surface of fiber) as the surface chemistry will also determine the nature of bonds formed between the coating layer and fiber and dictate the reversibility of the process.

4.2. Designing

The process and approach to designing are quite different from the traditional method of printing electrical components onto a circuit board. This variation is attributed to the versatility of textiles. As explained above, it is a very application-based approach.

The first step must be to realize what is the expected function of the smart textile and the degree of wearability (ranges from static to fully wearable). Based on step one, the substrate, fabric structure, sensing, and transducer-coating materials should be decided which becomes step 2. During the coating process, it is extremely crucial to obtain a uniform layer of coating. Cover factor C and power factor (PF) must be calculated according to Equations (1) and (2). According to Tracton and colleagues’ paper, the penetration of the coating into the fiber is calculated using $$\begin{array}{}\text{(1)}& C=w\ast d_w+f\ast d_f-w\ast f\ast d_w\ast d_f,\end{array}$$where $w$ is the number of warp threads per inch, $f$is the filling threads per inch, $d_w$ is the diameter of warp in inches, and $d_f$ is the diameter of the filling yarn in inches. $$\begin{array}{}\text{(2)}& \text{PF}=\frac{\text{Fabric}\text{Density}}{\text{Fiber}\text{Density}}=\frac{W\ast V_{\text{Fabric}}}{W\ast V_{\text{Fiber}}}=\frac{V_{\text{Fiber}}}{V_{\text{Fabric}}}\end{array}$$

In the scenario of nonuniform coating, micro-cracks formation probability increases, leading to overall deformations. Moreover, the optimal functioning of the sensor is inversely proportional to the nonuniformity of the coating. Therefore, proper adhesion of the coating, as well as its uniform thickness, plays a prime role. Mechanical, chemical, and plasma-based surface modifications can aid in ensuring the same. The research of Luo and Van Ooij is an excellent resource for learning more about surface modification techniques. In step 3, the chosen transducing material will determine the kind of power source and the system would need to function. In step 4, the data transferring system must be finalized. In addition, since the necessity for textile-based energy harvesters has been established in previous sections, step 5 is an extra step indicated by this study, which is to introduce connections between the parent system and an energy harvester that can aid in the constant supply of energy. Finally, in step 6, testing of the entire set-up must be performed.

4.3. Textile-Based Energy Harvesters

Textile-based energy harvesters provide numerous advantages over standard batteries in various biomedical devices already in use, including unlimited capacity, convenience, low maintenance, and no need for battery replacement. There are undeniably a slew of concerns at the individual, technical, and environmental levels that are driving the shift to textile-based energy harvesters. These energy harvesters make use of the energy that is wastefully released by the human body (mentioned in section III). On an average day, inclusive of all the activities performed by a normal individual, the biomechanical energy sums up to approximately 70 W. Thus, being able to utilize this and use it as a source of power for smart textiles rather than regular batteries is an intriguing, environmentally benign, and sustainable option that deserves careful consideration. Textile-based energy harvesters like smart textiles follow a similar process of fabrication. These harvesters can sense and transform energy from one form to another. They come in a variety of shapes and sizes, depending on the type of energy they convert or transform.

4.3.1. Triboelectric Energy Harvester

Mechanical energy expensed by the body is the leading source for the generation of electricity. It is an amalgamation of two processes: triboelectrification (resultant of the physical contact between two materials with varied electron affinity) and electrostatic induction (the potential difference that is created when two opposite charge carrying materials are separated). The shift towards miniature technology has shown property enhancement as the size of materials shifts from the macro-micro-nanoscale. This section will focus on energy harvesting triboelectric nanogenerators (TENGs). In 2012, Wang and his team introduced this concept and demonstrated its capabilities. Many notable research works have also showcased the advantages offered by TENGs [32–37]. TENGs can potentially harvest energy from sound, wind and vibrations. Furthermore, its fabrication involves materialistic diversity, thus aiding in choosing the material as specific as the application demands.

Zou et al. [38] have proved, and the TENGs can be made of materials such as cotton, wool, silk, artificial fiber, rubber, and polyurethane. As explained in Section 4, fabrics can be coated with various functional materials. Therefore, bridging the two concepts gives birth to a textile-based triboelectric energy harvester [39, 40]. A common concern with textile-based biomedical systems is washability. This concern was washed away by Ning et al. [41] who worked on a textile-based TENG that could generate a peak voltage of about 1050 V by measuring the friction within the clothes during everyday activities. They used nanofiber with high hydrophobic properties to impart the convenience of washing. The nanofiber was then woven into a nonwoven textile substrate via a shuttle flying process, and they have shown that a simple fabric (lab coat) can be incorporated with such a functionalized polymer fiber to generate electricity (Figure 5).

[figure(s) omitted; refer to PDF]

Their work intends to showcase the importance of bridging the merits of textile-based energy harvesters to solve the predicament faced by textile-based WBDs. In their work, they used electricity to power 54 LEDs. By replacing the LEDs with an inlet with a textile-based WBD system, we can have a sustainable long-lasting energy source that will ultimately enable continuous monitoring of vitals.

Zhou et al. [42] proved that using functional fiber inserted into textile could aid physiological monitoring using the power it generated in a newly published study. A combination of nylon fabric (positive) and polyester (negative) served as the source for triboelectrification with a conductive silver fiber fabric at the center. These textile-based TENGs developed by team Zhou et al. [43] were breathable, washable, and convenient, and they went on to produce alternating current by exploiting the energy expense at the arm joints during elbow movement. These composite fibers can feed in energy to textile-based WBDs, as shown in Figure 6.

[figure(s) omitted; refer to PDF]

When the principle behind the different forms of textile-based energy harvesting is understood, this enables the conjunction of two other energy harvesting systems as well. Interconnection between the two energy harvesting systems and material compatibility with textile manufacturing processes must be well understood before undertaking such a feat. To guide the reader towards understanding the same, we have highlighted the works of Chen et al. Chen et al. [44], in their work, explained the energy harvesting from not only mechanical sources but also solar at the same time. Another important highlight of their research was the interconnection technique used to connect the two different energy harvesters. They developed two modules, one being the module that harnesses energy from incident radiation of the sun, hence photovoltaic, and the second being the triboelectric/TENG, which harnesses power from the friction in the body motion, the action of wind blowing, vibrations, and deformations.

This was achieved using a polymer fiber-based solar cell and interlacing two polymer fibers with varied electron affinity to generate a triboelectrification effect. Another important aspect to note is that they ensured all the chosen polymer-fiber/electrodes were compatible for industrial textile processing. Figure 7 shows the interlayer distinction of one such fabric-based TENG in the form of a yarn.

[figure(s) omitted; refer to PDF]

Within the photovoltaic module, the ZnO/dye becomes the interface region where electrons and holes get separated due to the energy from incident sunlight, forming electron-hole pairs. The wire-shaped counter electrode acts as a collector for holes, while electrons move through the semiconductor’s conduction band. They are then collected at the photo anode, thus generating electricity from solar energy. Within the TENG module, the two PTFE strips are placed adjacently and interlaced with the copper electrode. When any deformations occur, mechanical excitation causes the two PTFE strips to come into contact, followed by deformation removal, resulting in a backflow of electrons between the electrodes, eventually leading to the current generation, as seen in Figure 8.

[figure(s) omitted; refer to PDF]

Both the independent modules were interconnected by counter electrodes, and copper was used as the common electrode. The interconnection procedures and materials used are explained in the same work. Moreover, the interlacing and other processes are also presented as visual aids/videos for the reader in their supplementary information. Thus, this work showcases the vast potential of such hybrid textiles possessing more than one energy harvesting system. The electricity generated from their hybrid textile was significant enough to charge a cell phone and an electronic watch under ambient conditions. If in a similar manner, this power generated could be connected to a textile-based WBD (parent system), this could lead to a fully self-powered textile-based biomedical device.

4.3.2. Piezoelectric Energy Harvesting

Piezoelectricity is the term for the generation of electricity by converting mechanical energy to electrical energy through the use of a specific material and motions involving any compression-based deformations (from the human body’s perspective), whether caused by movements, the wind, or being struck/squeezed. Thus, most wearable electronics possess a piezoelectric component within their design. Textile-based piezoelectric energy harvesting similarly will possess a patch of piezoelectric material at the location of the source of mechanical energy [45–50]. Just as TENGs for triboelectric generators, here for piezoelectric energy harvesting, we have piezoelectric nanogenerators (PENGs). Polyvinylidene fluoride (PVDF) is the most commonly utilized material component for textile-based PENG because it is easy to incorporate into a fabric substrate [51, 52].

For a personalized health monitoring system, such a fabric-based patch can be directly attached or incorporated into the textile, with the help of interconnections, and the energy generated from the PENG module can be utilized to power the textile-based WBD (parent system) [53, 54]. Qin et al. [55] developed ZnO-based PENG on a Kevlar fiber by hydrothermal process. Since the fiber is coated with ZnO utilizing a fiber and twisted layer, the force created by stretching can be determined as illustrated in Figure 9. Tetraethoxysilane was utilized as a mechanical strengthening component, and the whole structure generated a voltage of 2.2 V with a current of 60-120 nA. This fabric-based PENG could be used to detect heartbeat and extract energy from it to create power under low-frequency biomechanical energy conditions.

[figure(s) omitted; refer to PDF]

Yang et al. [56], in their works, explored a similar application of employing ZnO nanowires for textile-based PENG. They were able to produce an oscillating voltage up to 50 mV during the action of stretching and releasing. An ultrahigh piezoelectric voltage of 0.79 Vm $N-1$ was developed by Chen and his team. They synthesized lead zirconate titanate (PZT) fibers via sol-gel technique, and a layer of polydimethylsiloxane (PDMS) was coated on it. On the application of pressure, the strain created flowed through the PZT fiber and passed the PDMS to generate charges during their testing. They suggested that by using more efficient piezoelectric material to coat a fiber substrate, the output voltage of 1.63 V that was observed could be improved. Further to harness the energy from a human arm, Zhang et al. developed a woven fabric generator of BaTiO₃-PVC composite fiber, which produced a voltage of 1.9 V. Gu et al. [57] further added to this research where they explained vertically aligned PENGs yield more voltage. In their work, a voltage of around 209 V was generated. This was a huge leap in the PENG research. Such high output, if interconnected to a textile-based WBD, could promote the shift towards a self-powering health monitoring system.

Thus, it is logical to say that in order to produce flexible textile-based PNG, the two electrodes used for PENG must be flexible and fibrous, or else attaching to the human body may not be a comfortable fit. We would like to highlight the work of Ji et al. [58] and the team, as they developed a yarn-based PENG in the form of a core-shell structure. There were two conductive threads used, an inner layer (inner electrode) and an outer layer (outer electrode). The piezoelectric nanofiber BNT-ST/PVDF was wound around the inner electrode. The core-shelled piezoelectric yarn/nanofiber was braided with two strands of conductive thread (outer electrode). For interested readers who wish to understand the synthesis process of BNT-ST/PVDF nanofibers, it has been elaborately explained in Ji et al.’s works.

This specific structural integration of components leads to improved performance and durability as it is sensitive towards bodily deformations. The core-shelled piezoelectric yarn, which was braided with the electrodes, was then directly stitched onto fabric to yield the final textile-based PENG. When this textile-based PENG was tested, the output voltage was found to be 50 kPa. Their work investigated that the voltage variation is based on the length of the PENG patch stitched, the stitching interval between the yarns and the pattern of stitching. Thus, these factors must be considered as prime factors to evaluate the voltage output with precision. They are able to introduce this patch in gloves and elbow region as a guard, and a considerable amount of power can be generated as a result of stretching and bending of these parts as illustrated in Figure 10.

[figure(s) omitted; refer to PDF]

Thus, if a shirt with textile-based WBD is developed, it could incorporate these PENG patches at the source of deformations, with proper interconnections between different modules, as shown by Chen et al. in their study. It may self-generate the power required to run the device without the need for batteries to be replaced.

4.3.3. Thermoelectric Energy Harvesters

Thermoelectric energy harvesting systems have gained attention primarily because of their working principle. It functions based on the heat released by the human body and the temperature gradient of the environment. It requires no additional deformations or body movements or external sources of energy such as sunlight. A thermoelectric energy harvester generally possesses two sides: cold and hot (refer to Figure 11). The hot side, which is between the rough skin surface and smooth thermoelectric element, generates thermal contact resistance. Subsequently, on the cold side, the heat is released via convection. Normally, this region needs to have a high surface area for higher convection. To incorporate the thermoelectric materials onto textile, a preferred polymer is poly (3, 4- ethylenedioxythiophene): poly (styrene sulfonate) (PEDOT: PSS). This is a type of conductive polymer complex. The main advantage of this polymer complex is the ease with which it can be transformed into a full-fledged textile-based system. Coating and spinning are the two main methods that facilitate this incorporation.

[figure(s) omitted; refer to PDF]

In previous research works, both in-plane and out-of-plane devices based on thermoelectric polymer-based textiles have been explored. Du et al. [59] coated cotton with PEDOT: PSS and used a silver wire. The cotton strip was $35\text{mm}\text{x}5\text{mm}$, and it produced a maximum power output of 12.16 nW. Using vapor deposition, p-doped PEDOT-Cl was deposited onto cotton for an out-of-plane thermoelectric device by Allison et al. [60]. The power generated was around 4.5 nW.

In the most recent research work, Lund et al. [61] and the team have developed eight thermocouples whose each leg pair generates 2.3 nW per degree at a gradient of temperature 65 K with an overall power output of 1.2 microwatts. The highlight of this research is not only the high power output but the compatibility of their textile-based thermoelectric material with the existing textile manufacturing processes for large-scale applications. This group has also created plain and dyed conductive polymer rolls that are wash and wear-resistant and ready to be woven and sewn.

Another intriguing finding from their research is the use of conducting paste as an interface between thermocouples to lower component contact resistance. To generate the out-of-plane layout, they perform embroidery by hand to hold the different layers of wool fabric in place, as shown in Figure 12. Wool fabric was chosen for its comfort and ability to generate a temperature gradient. Therefore, when this textile-based thermoelectric energy harvesting system is worn in low-temperature conditions, the body becomes the hot region, and the external environment becomes the cold; thus, the heat will flow from hot to cold region, which results in the production of electricity, as shown in Figure 12. With such research happening, in our review, we would like to highlight once again the importance of creating a link between textile-based energy harvesters and textile-based biomedical wearable devices. This could revolutionize the field of health monitoring systems forever, as the main drawback of traditional biomedical devices can be addressed.

[figure(s) omitted; refer to PDF]

5. Conclusion

An increasingly health-conscious and convenience-driven culture has prompted the convergence of technology and medicine. As a result, a variety of invasive and noninvasive wearable devices that can capture biosystemic data have been developed, and they have proven to be useful in disease diagnosis and detection. However, their effective acceptance by wearers/patients has been hampered by their performance life-span, wearability, expense, and comfort. Furthermore, its psychological impact has been identified as a contributing factor in slow recovery rates. Fortunately, with the advancement in miniature technology and the textile industry, this new fusion can be the solution to the predicaments faced by traditional wearable devices. The solution lies in bridging textile-based wearable biomedical devices with textile-based energy harvesters which can provide the continuous source of power that can be generated from the wasteful energy released by various human motions using source-specific textile-based nanogenerators. Therefore, we hope to plant a seed of thought in the minds of the readers and researchers about the importance of textile-based energy harvesters and their vast but less explored potential.

Acknowledgments

The authors would like to thank the Innovation in Science Pursuit for Inspired Research (INSPIRE) Faculty Program through the Department of Science and Technology (DST) funded by the Ministry of Science and Technology (DST/INSPIRE/04/2017/002629).

References

[1] S. Park, S. Jayaraman, "Smart textile-based wearable biomedical systems: a transition plan for research to reality," IEEE Transactions on Information Technology in Biomedicine, vol. 14 no. 1, pp. 86-92, DOI: 10.1109/titb.2009.2025817, 2010.

[2] M. Anderson, "What are the components of a patient monitoring system? ATL Technology," 2020. https://atltechnology.com/blog/components-of-patient-monitoring-systems/

[3] Y. Zhang, J. Ding, B. Qi, W. Tao, J. Wang, C. Zhao, H. Peng, J. Shi, "Multifunctional fibers to shape future biomedical devices," Advanced Functional Materials, vol. 29 no. 34,DOI: 10.1002/adfm.201902834, 2019.

[4] S. M. A. Iqbal, I. Mahgoub, E. Du, M. A. Leavitt, W. Asghar, "Advances in healthcare wearable devices," NPJ Flexible Electronics, vol. 5 no. 1,DOI: 10.1038/s41528-021-00107-x, 2021.

[5] S. J. Benight, C. Wang, J. B. H. Tok, Z. Bao, "Stretchable and self-healing polymers and devices for electronic skin," Progress in Polymer Science, vol. 38 no. 12, pp. 1961-1977, DOI: 10.1016/j.progpolymsci.2013.08.001, 2013.

[6] A. dos Santos, E. Fortunato, R. Martins, H. Águas, R. Igreja, "Transduction mechanisms, micro-structuring techniques, and applications of electronic skin pressure sensors: a review of recent advances," Sensors, vol. 20 no. 16,DOI: 10.3390/s20164407, 2020.

[7] H.-H. Chou, A. Nguyen, A. Chortos, C. Lu, J. Mei, T. Kurosawa, W.-G. Bae, J. B.-H. Tok, Z. Bao, "A chameleon-inspired stretchable electronic skin with interactive colour changing controlled by tactile sensing," Nature Communications, vol. 6 no. 1,DOI: 10.1038/ncomms9011, 2015.

[8] T. Someya, M. Amagai, "Toward a new generation of smart skins," Nature Biotechnology, vol. 37 no. 4, pp. 382-388, DOI: 10.1038/s41587-019-0079-1, 2019.

[9] J. Mark, Polymer Data Handbook, 1999.

[10] M. C. Bélanger, Y. Marois, "Hemocompatibility, biocompatibility, inflammatory and in vivo studies of primary reference materials low-density polyethylene and polydimethylsiloxane: a review," Journal of Biomedical Materials Research, vol. 58 no. 5, pp. 467-477, DOI: 10.1002/jbm.1043, 2001.

[11] B. A. Grzybowski, S. T. Brittain, G. M. Whitesides, "Thermally actuated interferometric sensors based on the thermal expansion of transparent elastomeric media," The Review of Scientific Instruments, vol. 70 no. 4, pp. 2031-2037, DOI: 10.1063/1.1149706, 1999.

[12] T. N. Tran, J. A. Heredia-Guerrero, B. T. Mai, L. Ceseracciu, L. Marini, A. Athanassiou, I. S. Bayer, "Bioelastomers based on cocoa shell waste with antioxidant ability," Advanced Sustainable Systems, vol. 1 no. 7, article 1700002,DOI: 10.1002/adsu.201700002, 2017.

[13] D. Sun, Characterization of Medical Grade Poly-Dimethylsiloxane as Encapsulation Materials for Implantable Microelectromechanical Systems, 2014.

[14] A. Nathan, A. Ahnood, M. T. Cole, S. Lee, Y. Suzuki, P. Hiralal, F. Bonaccorso, T. Hasan, L. Garcia-Gancedo, A. Dyadyusha, S. Haque, "Flexible electronics: the next ubiquitous platform," Proceedings of the IEEE, vol. 100, pp. 1486-1517, DOI: 10.1109/JPROC.2012.2190168, 2012.

[15] J.-g. Liu, H.-j. Ni, Z.-h. Wang, S.-y. Yang, W.-f. Zhou, "Colorless and transparent high–temperature-resistant polymer optical films–current status and potential applications in optoelectronic fabrications, in: optoelectronics-materials and devices," Optoelectronics - Materials and Devices,DOI: 10.5772/60432, 2015.

[16] M. Kaltenbrunner, T. Sekitani, J. Reeder, T. Yokota, K. Kuribara, T. Tokuhara, M. Drack, R. Schwödiauer, I. Graz, S. Bauer-Gogonea, S. Bauer, T. Someya, "An ultra-lightweight design for imperceptible plastic electronics," Nature, vol. 499 no. 7459, pp. 458-463, DOI: 10.1038/nature12314, 2013.

[17] C. J. Lee, S. J. Oh, J. K. Song, S. J. Kim, "Neural signal recording using microelectrode arrays fabricated on liquid crystal polymer material," Materials Science and Engineering C, vol. 24 no. 1-2, pp. 265-268, DOI: 10.1016/j.msec.2003.09.143, 2004.

[18] M. Irimia-Vladu, P. A. Troshin, M. Reisinger, G. Schwabegger, M. Ullah, R. Schwoediauer, A. Mumyatov, M. Bodea, J. W. Fergus, V. F. Razumov, "Environmentally sustainable organic field effect transistors," Organic Electronics, vol. 11 no. 12, pp. 1974-1990, DOI: 10.1016/j.orgel.2010.09.007, 2010.

[19] P. Boonvisut, M. C. Çavuşoğlu, "Estimation of soft tissue mechanical parameters from robotic manipulation data," IEEE/ASME Transactions on Mechatronics, vol. 18 no. 5, pp. 1602-1611, DOI: 10.1109/TMECH.2012.2209673, 2013.

[20] E. Series, Super-soft, addition cure silicone rubbers, smooth-on, 2015.

[21] T. R. Ray, J. Choi, A. J. Bandodkar, S. Krishnan, P. Gutruf, L. Tian, R. Ghaffari, J. A. Rogers, "Bio-integrated wearable systems: a comprehensive review," Chemical Reviews, vol. 119 no. 8, pp. 5461-5533, DOI: 10.1021/acs.chemrev.8b00573, 2019.

[22] R. Riemer, A. Shapiro, "Biomechanical energy harvesting from human motion: theory, state of the art, design guidelines, and future directions," Journal of Neuroengineering and Rehabilitation, vol. 8 no. 1,DOI: 10.1186/1743-0003-8-22, 2011.

[23] L. C. Rome, L. Flynn, E. M. Goldman, T. D. Yoo, "Generating electricity while walking with loads," Science, vol. 309 no. 5741, pp. 1725-1728, DOI: 10.1126/science.1111063, 2005.

[24] D. Vélez Día, S. S. Moreno Gutiérrez, "Biomechanics and motor control of human movement," XIKUA Boletín Científico de La Escuela Superior de Tlahuelilpan, vol. 1 no. 1,DOI: 10.29057/xikua.v1i1.1175, 2013.

[25] J. Ortiz, Wearable technologies, 2018.

[26] R. M. Alexander, "Biomechanics and motor control of human movement. David A. Winter," The Quarterly Review of Biology, vol. 66 no. 1, pp. 106-106, DOI: 10.1086/417116, 1991.

[27] A. D. Winter, E. A. Patla, S. J. Frank, E. S. Walt, "Biomechanical walking pattern changes in the fit and healthy elderly," Physical Therapy, vol. 70 no. 6, pp. 340-347, DOI: 10.1093/ptj/70.6.340, 1990.

[28] M. R. Shorten, "The energetics of running and running shoes," Journal of Biomechanics, vol. 26, pp. 41-51, DOI: 10.1016/0021-9290(93)90078-s, 1993.

[29] L. M. Castano, A. B. Flatau, "Smart fabric sensors and e-textile technologies: a review," Smart Materials and Structures, vol. 23 no. 5, article 053001, 2014.

[30] A. A. Tracton, Coatings Technology Handbook,DOI: 10.1201/9781420044089, 2006.

[31] S. Luo, W. J. Van Ooij, "Surface modification of textile fibers for improvement of adhesion to polymeric matrices: a review," Journal of Adhesion Science and Technology, vol. 16 no. 13, pp. 1715-1735, DOI: 10.1163/156856102320396102, 2002.

[32] Z. L. Wang, L. Lin, J. Chen, S. Niu, Y. Zi, "Harvesting body motion energy," Triboelectric Nanogenerators, pp. 207-236, 2016.

[33] J. Y. Park, M. Salauddin, M. S. Rasel, "Nanogenerator for scavenging low frequency vibrations," Journal of Micromechanics and Microengineering, vol. 29, article 053001, 2019.

[34] F. Yi, Z. Zhang, Z. Kang, Q. L. Liao, Y. Zhang, "Recent advances in triboelectric nanogenerator-based health monitoring," Advanced Functional Materials, vol. 29 no. 41, article 1808849,DOI: 10.1002/adfm.201808849, 2019.

[35] F. Invernizzi, S. Dulio, M. Patrini, G. Guizzetti, P. Mustarelli, "Energy harvesting from human motion: materials and techniques," Chemical Society Reviews, vol. 45 no. 20, pp. 5455-5473, DOI: 10.1039/C5CS00812C, 2016.

[36] Z. L. Wang, "Triboelectric nanogenerators as new energy technology for selfpowered systems and as active mechanical and chemical sensors," ACS Nano, vol. 7 no. 11, pp. 9533-9557, DOI: 10.1021/nn404614z, 2013.

[37] J. Chen, Triboelectric Nanogenerators, 2016. http://hdl.handle.net/1853/54956

[38] Y. Zou, V. Raveendran, J. Chen, "Wearable triboelectric nanogenerators for biomechanical energy harvesting," Nano Energy, vol. 77,DOI: 10.1016/j.nanoen.2020.105303, 2020.

[39] K. Parida, G. Thangavel, G. Cai, X. Zhou, S. Park, J. Xiong, P. S. Lee, "Extremely stretchable and self-healing conductor based on thermoplastic elastomer for all-three-dimensional printed triboelectric nanogenerator," Nature Communications, vol. 10 no. 1,DOI: 10.1038/s41467-019-10061-y, 2019.

[40] T. He, Q. Shi, H. Wang, F. Wen, T. Chen, J. Ouyang, C. Lee, "Beyond energy harvesting-multi-functional triboelectric nanosensors on a textile," Nano Energy, vol. 57, pp. 338-352, DOI: 10.1016/j.nanoen.2018.12.032, 2019.

[41] C. Ning, L. Tian, X. Zhao, S. Xiang, Y. Tang, E. Liang, Y. Mao, "Washable textile-structured single-electrode triboelectric nanogenerator for self-powered wearable electronics," Journal of Materials Chemistry A, vol. 6 no. 39, pp. 19143-19150, DOI: 10.1039/c8ta07784c, 2018.

[42] Z. Zhou, S. Padgett, Z. Cai, G. Conta, Y. Wu, Q. He, S. Zhang, C. Sun, J. Liu, E. Fan, K. Meng, Z. Lin, C. Uy, J. Yang, J. Chen, "Single-layered ultra-soft washable smart textiles for all-around ballistocardiograph, respiration, and posture monitoring during sleep," Biosensors and Bioelectronics, vol. 155,DOI: 10.1016/j.bios.2020.112064, 2020.

[43] T. Zhou, C. Zhang, C. B. Han, F. R. Fan, W. Tang, Z. L. Wang, "Woven structured triboelectric nanogenerator for wearable devices," ACS Applied Materials & Interfaces, vol. 6 no. 16, pp. 14695-14701, DOI: 10.1021/am504110u, 2014.

[44] J. Chen, Y. Huang, N. Zhang, H. Zou, R. Liu, C. Tao, X. Fan, Z. L. Wang, "Micro-cable structured textile for simultaneously harvesting solar and mechanical energy," Nature Energy, vol. 1 no. 10,DOI: 10.1038/nenergy.2016.138, 2016.

[45] W. Deng, T. Yang, L. Jin, C. Yan, H. Huang, X. Chu, Z. Wang, D. Xiong, G. Tian, Y. Gao, H. Zhang, W. Yang, "Cowpea-structured PVDF/ZnO nanofibers based flexible self-powered piezoelectric bending motion sensor towards remote control of gestures," Nano Energy, vol. 55, pp. 516-525, DOI: 10.1016/j.nanoen.2018.10.049, 2019.

[46] B. Yang, K.-S. Yun, "Piezoelectric shell structures as wearable energy harvesters for effective power generation at low-frequency movement," Sensors and Actuators A: Physical, vol. 188, pp. 427-433, DOI: 10.1016/j.sna.2012.03.026, 2012.

[47] E. Nilsson, L. Mateu, P. Spies, B. Hagström, "Energy harvesting from piezoelectric textile fibers," Process Engineering, vol. 87, pp. 1569-1572, DOI: 10.1016/j.proeng.2014.11.600, 2014.

[48] D. Matsouka, S. Vassiliadis, D. V. Bayramol, "Piezoelectric textile fibres for wearable energy harvesting systems," Materials Research Express, vol. 5 no. 6, article 065508,DOI: 10.1088/2053-1591/aac928, 2018.

[49] E. Yang, Z. Xu, L. K. Chur, A. Behroozfar, M. Baniasadi, S. Moreno, J. Huang, J. Gilligan, M. Minary-Jolandan, "Nanofibrous smart fabrics from twisted yarns of electrospun piezopolymer," ACS Applied Materials & Interfaces, vol. 9 no. 28, pp. 24220-24229, DOI: 10.1021/acsami.7b06032, 2017.

[50] W. Zeng, X.-M. Tao, S. Chen, S. Shang, H. L. W. Chan, S. H. Choy, "Highly durable all-fiber nanogenerator for mechanical energy harvesting," Energy & Environmental Science, vol. 6 no. 9,DOI: 10.1039/c3ee41063c, 2013.

[51] F. Mokhtari, G. M. Spinks, C. Fay, Z. Cheng, R. Raad, J. Xi, J. Foroughi, "Wearable electronic textiles from nanostructured piezoelectric fibers," Advanced Materials Technologies, vol. 5 no. 4,DOI: 10.1002/admt.201900900, 2020.

[52] F. Mokhtari, Z. Cheng, R. Raad, J. Xi, J. Foroughi, "Piezofibers to smart textiles: a review on recent advances and future outlook for wearable technology," Journal of Materials Chemistry A, vol. 8 no. 19, pp. 9496-9522, DOI: 10.1039/D0TA00227E, 2020.

[53] Z. L. Wang, "From nanogenerators to piezotronics—A decade-long study of ZnO nanostructures," MRS Bulletin, vol. 37 no. 9, pp. 814-827, DOI: 10.1557/mrs.2012.186, 2012.

[54] Z. L. Wang, J. Song, "Piezoelectric nanogenerators based on zinc oxide nanowire arrays," Science, vol. 312 no. 5771, pp. 242-246, DOI: 10.1126/science.1124005, 2006.

[55] Y. Qin, X. Wang, Z. L. Wang, "Microfibre-nanowire hybrid structure for energy scavenging," Nature, vol. 451 no. 7180, pp. 809-813, DOI: 10.1038/nature06601, 2008.

[56] R. Yang, Y. Qin, L. Dai, Z. L. Wang, "Power generation with laterally packaged piezoelectric fine wires," Nature Nanotechnology, vol. 4 no. 1, pp. 34-39, DOI: 10.1038/nnano.2008.314, 2009.

[57] L. Gu, N. Cui, L. Cheng, Q. Xu, S. Bai, M. Yuan, W. Wu, J. Liu, Y. Zhao, F. Ma, Y. Qin, Z. L. Wang, "Flexible fiber nanogenerator with 209 V output voltage directly powers a light-emitting diode," Nano Letters, vol. 13 no. 1, pp. 91-94, DOI: 10.1021/nl303539c, 2013.

[58] S. H. Ji, Y.-S. Cho, J. S. Yun, "Wearable core-shell piezoelectric nanofiber yarns for body movement energy harvesting," Nanomaterials, vol. 9 no. 4,DOI: 10.3390/nano9040555, 2019.

[59] Y. Du, K. F. Cai, S. Z. Shen, R. Donelsonand, J. Y. Xu, H. X. Wang, T. Lin, "Multifold enhancement of the output power of flexible thermoelectric generators made from cotton fabrics coated with conducting polymer," RSC Advances, vol. 7 no. 69, pp. 43737-43742, DOI: 10.1039/c7ra08663f, 2017.

[60] L. K. Allison, T. L. Andrew, "A wearable all-fabric thermoelectric generator," Advanced Materials Technologies, vol. 4 no. 5,DOI: 10.1002/admt.201800615, 2019.

[61] A. Lund, Y. Tian, S. Darabi, C. Müller, "A polymer-based textile thermoelectric generator for wearable energy harvesting," Journal of Power Sources, vol. 480, article 228836,DOI: 10.1016/j.jpowsour.2020.228836, 2020.