Discharge characteristics of poly(3,4-ethylene dioxythiophene): poly(styrenesulfonate) (PEDOT:PSS) textile batteries;comparison of silver coated yarn electrode devices and pure stainless steel filament yarn electrode devices

Abstract

This paper investigates textile-based energy storage devices fabricated with poly(3,4-ethylene dioxythiophene):poly(styrenesulfonate) (PEDOT:PSS) as an electro-active polymer and conductive yarns as the electrodes. The conductive yarns are sewn into a textile substrate and then coated with PEDOT:PSS systematically. Two different sets of devices were made. A comparison of the devices made with silver coated polybenzoxazol filament yarns and the devices made with pure stainless steel filament yarns is performed. The devices were charged and their self-discharge was measured by voltage decay. A study of the influence of charging time on the decay and the effect brought by various load resistors on the voltage decay is also performed. In this research, the devices with electrodes of pure stainless steel filaments yarns performed better than the devices with silver coated yarns; this outcome has been reported as standard by various researchers.

Keywords

coating conductive yarns diffusion electro-active polymer PEDOT:PSS textile battery

Textile-based electronic devices like transistors, light emitting diodes, biosensors, super capacitors and energy storage devices are made from electro-active polymers in combination with conductive yarns under special fabrication methods.^1–4 These are incorporated or integrated into traditional textile substrates to produce high innovative products that are light weight. These innovative products are referred to as smart textiles or electronic textiles^3,5 in some literature. Smart textiles are able to respond to environmental stimuli while the e-textiles are obtained by integrating electronic components into a textile substrate. Conducting polymers such as poly(3,4-ethylene dioxythiophene):poly(styrenesulfonate) (PEDOT(polyethylene dioxythiophene)) are playing an important role as organic semiconductors in electronic applications especially when they are p- and n- dopable, since they can transport either holes or electrons.⁶

A complete smart textile system would consist of a sensor, actuator, data processing, energy storage device and links connecting these devices. Most of the existing smart textile prototypes have been produced by integrating bulky, rigid and heavy electronic devices into flexible textiles. This reduces the comfort of the clothing and increases the weight; hence, there is a need to fully embed these devices into the textile structure without compromising the comfort and other desirable aspects of the clothing.

So far, various devices of the smart textile system, i.e. actuators, sensors and interconnects, have been fabricated from conductive yarns⁷ and electro-active polymers.¹ However, a textile-based battery/energy storage device integrated into the textile structure that is thin, flexible and compatible with other textile aspects is still a challenge to produce.

A battery is an electrochemical cell that converts stored chemical energy into electrical energy.⁸ It supplies the required energy to drive the system. A conventional battery consists of a cathode as the positive electrode and an anode as the negative electrode. It also has an electrolyte which normally undergoes electrochemical reaction, and makes the cations and anions move towards the cathode and the anode respectively thus conducting current. A separator is also a main component of the battery that separates the anode and the cathode. Some conventional batteries also have current collectors that are connected to both the cathode and the anode.

Remarkable success has been achieved by producing energy storage devices through incorporating single walled carbon nanotubes (SWCN) into textiles, and also using graphite electrodes.^9,10 However, these compounds are not very safe to human health and to the environment. Flexible and stretchable film batteries have been made using lithium battery materials woven with the yarns to make batteries for smart textile applications.¹¹

The electro-active polymer, PEDOT:PSS which has been chosen for this research has greater flexibility with respect to its chemistry and physics. It has been used both as an electrode material (spin coated) and as an electrolyte.¹² It is a safer polymer to work with compared to carbon based nano materials that are used by other researchers. The polymer can withstand 200℃ without degradation; therefore, it has superior thermal properties in ambient air. It also has good electrochemical stability, charge capacity and ionic conductivity.^13–15 Electro conductive stainless steel yarns are reported to be used in production of textile electrodes, electromagnetic shielding fabrics and smart garments for healthcare and other areas. They have the advantage of flexibility, comfortability and the ability to be integrated into the fabrics by stitching, knitting or weaving.^16–18 On the other hand, they have good electrical and thermal conductivity with high melting points. Silver coated yarns have also been used, but they are more expensive than stainless steel yarns. In this work, pure stainless steel yarns and silver coated yarns are used as ‘bipolar’ electrodes in the fabrication of the energy storage device.

Bhattacharya et al.¹² made a rechargeable textile battery using PEDOT:PSS as an active electronic coating and silver coated polyamide yarns as electrodes, on a textile substrate. However, it is shown in this research that the combination of PEDOT:PSS and stainless steel yarns used as electrodes gives better results compared to silver coated polybenzoxazol filament yarns.

Methodology

Device preparation

Available polyamide, cotton/polyester and pure polyester fabrics were tested to choose the appropriate textile substrate. We settled for cotton/polyester because of the fabric structure and good wettability of the cotton counterpart. Therefore, the PEDOT:PSS polymer penetrated well into the fabric structure. A three layered porous, laminate of textile substrate (5 cm by 5 cm) was made with this fabric, and an adhesive interlining was used to attach the different layers to each other. The fabric structure used was a 3/1 twill weave, with weft density of 18 yarns/cm and warp density of 39 yarns/cm. The yarn count was 19.5 tex for warp 48 tex for weft. Three pairs of conductive yarns (electrodes), approximately 6 cm in length were sewn into the surface fabric, 1 mm apart before lamination. The upper surface of the fabric was made hydrophobic by using a thermoplastic polyurethane (TPU) layer, except for a center region of 10 mm by 6 mm. PEDOT:PSS from Ossila was coated in layers (seven layers, to increase the coating uniformity and to reduce the bulk resistance) on the area that was not covered with TPU. The PEDOT:PSS (approximately 0.5 ml) was dropped in with a syringe uniformly on the fabric while it was in the oven. Each layer of PEDOT:PSS was left to cure in the oven for 15 minutes at a temperature of 90–100℃, before applying the next layer. The produced samples are as presented in Figure 1. Therefore, two types of conductive textile yarns were used to produce two different sets of energy storage devices with silver coated polybenzoxazol (PBO) filament yarns from AmberStrand® fiber and pure stainless steel filament yarn from Bekinox®Bekaert. Copper coated PBO (metal clad polymer) was also tested as an electrode. Current conductivity between electrodes during charging process was possible, but the charge storage performance was poor, hence, not reported on this paper. A schematic view is shown in Figure 2(a) for the top view of the device, and Figure 2(b) is the cross-sectional view of the device along line AA’ shown in Figure 2(a). Although two electrodes are sufficient, three were sewn in order to carry out the experiments in case one contact failed. Neither the PEDOT:PSS layer nor the entire cell was covered by the protecting layer and were permanently exposed to the ambient environment.
Figure 1.
Pure stainless steel filament yarns electrode sample and silver coated PBO filament yarns electrode sample.

Figure 2.
Schematic: (a) top view of the PEDOT:PSS device and (b) cross-sectional view of the PEDOT:PSS device along AA’.

One may wonder why the word ‘battery’ and not ‘capacitor’ has been used here. A battery converts chemical energy into electric energy. Electrochemical reactions are taking place at both electrodes. In our case, two electrodes made from the same material are being used. Strictly speaking we are then dealing with a capacitor. On the other hand we cannot exclude that some electrochemical reactions are taking place, because the physical mechanism of charge storage in PEDOT:PSS is still not well understood. We used the word ‘battery’ because it was also proposed by other authors.¹² As a consequence, one may not denote the electrodes as a cathode or an anode, because they are both made from the same material. Moreover, the PEDOT:PSS device has no predefined polarity. Both electrodes can be used for positive or negative voltages.

Charging and discharging procedure

The samples were charged using a power supply PL601 always at a constant voltage of 1.5 V. The charging time t_ch was varied according to the experiment in question (varying between 5 minutes and 240 minutes). For experiments with a load resistor R, the load was connected in parallel, as shown in Figure 3. Charge decay measurements were taken from a multimeter Digitool Digi 16, recorded by a digital camera. The voltage decay measurement could be repeated up to three times on the same device with no significant change in the discharge profile.
Figure 3.
Schematic lay-out of experimental set up.

Results

Experimental results without an external load resistor

First the sample voltages were measured before charging and the devices were found with no stored charge. PEDOT:PSS samples with silver coated yarn electrodes were charged at a constant voltage of 1.5 V for a duration of time t_ch. A voltage meter with a high input impedance of 10 mohm was used to record the voltage V across the sample after the power supply was switched off by opening the switch S at the initial time t = 0 (Figure 3).

The results obtained with the silver coated yarn electrode samples are shown in Figure 4 while those with pure stainless steel filament yarns are shown in Figure 5. Figure 4(a) shows the behavior of the voltage decay in the first 5000 seconds and Figure 4(b) shows the decay over a longer period of up to 15,000 seconds. A first observation is the sharp decrease of the voltage V in the beginning of the discharging process, as soon as the switch S is opened. But after some time, in the range of 1000 seconds, the voltage V decreases very slowly.
Figure 4.
Voltage decay for silver coated PBO filament yarns device for various charging times: (a) the decay behavior for the first 5000 seconds and (b) the decay behavior from zero to 15,000 seconds.

Figure 5.
Voltage decay for pure stainless steel filament yarns device for various charging times: (a) the decay behavior for the first 5000 seconds and (b) the decay behavior from zero to 15,000 seconds.

A second observation when studying the graphs is the influence of the charging time t_ch. The longer the charging time the slower the voltage decay and the higher the output voltage in the device. All curves show a similar voltage decay behavior. These devices too can be cycled a number of times (up to 10 times), comparable to the number of cycles reported for solid electrolyte-based lithium batteries obtained in the charge-discharge measurements by Liu et al.¹¹ and the cycling voltammetry measurements by Bhattacharya et al.¹²

In Figure 5, similar results can also be drawn for the samples made with pure stainless steel filament yarns as electrodes. The behavior of the output voltage is similar to the silver coated PBO filament yarn electrodes device, but the voltage values are higher. Compared to the silver coated PBO filament yarn electrodes whose decay tends to remain constant at 0.2 V, the curves for the steel yarn electrodes remain at a constant voltage of 0.4 V for quite a long time. These devices can be used for voltage stabilization during one hour if the load resistor is not too small.

However, after a very long time, when all the ions have reached their original position, the discharge voltage will go to zero anyway. The charging time also has a similar influence on the voltage decay as in the silver electrodes samples.

It is clear that the discharge of all the cells is very fast in the beginning just after switching off the power supply. After some time the discharge proceeds very slowly. For the stainless steel electrodes, the voltage drops from 1.5 V to 0.4 V after one hour (see Figure 6(b)), which is a rather low efficiency if one would use the cell for electric energy storage. The fast transient in the very beginning is typical for a diffusion process and probably this fact is responsible for the low efficiency of the actual cells. It must be pointed out here that the basic mechanisms occurring in PEDOT:PSS are still not completely known. The cells with stainless steel were found to be twice as good compared to the cell with the silver coated yarn electrodes. The difference between stainless steel yarns and silver coated yarns could be explained by the electrolytic phenomena observed by Bhattacharya et al.,¹² by electron microscope measurement; they clearly observed migration of silver particles.
Figure 6.
Comparison of voltage decay for a pure stainless steel filament yarn device and a silver coated PBO filament yarn device for different charging times: (a) five minutes of charging and (b) 120 minutes of charging.

Comparing both sets of devices as in Figure 6, it is clear that the stainless steel yarn gives better results. The device can store its charge for a longer period and with higher voltage values than the device with the silver coated yarns as electrodes. A similar conclusion of good performance of the pure stainless steel filament yarn textile electrodes has also been reported in medical application.¹⁶

Experimental results with a load resistor

From the experiments done so far, it is obvious that the PEDOT:PSS is self-discharging as the input impedance of the voltage meter is very high. In order to quantify this self-discharge, resistors R (see Figure 3) of 978 kohm, 268 kohm and 100 kohm were connected each at a time in parallel with the PEDOT:PSS capacitor. After charging for 120 minutes at a constant voltage of 1.5 V, the decaying voltage V was recorded. The obtained curves from this data are shown in Figure 7. It can be observed that the decay is faster, but with lower values, at the initial phase of the curve since it was without the resistor. It can also be observed that the lower the load resistor R, the faster the charge decay. This means that the device can only power high load resistors which require very little current and therefore can be used for voltage stabilization if the resistor is not too small.
Figure 7.
Result with load resistors for pure stainless steel filament yarn electrodes: (a) the voltage decay behavior for the first 5000 seconds and (b) the voltage decay behavior from zero to 10,000 seconds.

Experiments with a load resistor were performed with the silver electrodes device as well. However, these results are not shown, because these specific devices could barely support a load resistor.

In all the experiments completed so far, the voltage meter with its input resistance of 10 mohm was connected to the samples. One may wonder if we are dealing with a self-discharge or a discharge through the 10 mohm input resistance. Therefore, different discharge experiment was set up where the voltage meter was disconnected regularly for periods of five minutes. The meter was only connected for a short time just enough to measure the voltage. These curves turned out to be almost coinciding with the curves shown in Figure 5. The conclusion is that we were really measuring the self-discharge of a PEDOT:PSS cell. As the voltage meter has a negligible influence, it can be stated that the PEDOT:PSS cell has itself an internal resistance much lower than 10 mohm. From the discharge curves obtained with the different values of the load resistor, the internal resistance of the PEDOT:PSS cell could be estimated to be around 300 kohm. From most of the curves, one can observe time constants in the order of one hour or 3600 s. If the cell would be considered as a capacitor C connected in parallel with a resistance R one has RC = 3600 s from which we get that the capacitance is 18000 µF. This is quite a high value taking into account the limited area of the electrodes in contact with the dielectric. We must conclude then that only electrolytic phenomena could be responsible for such a high value, i.e. mobile ions move under the influence of the applied electric field. Cyclic voltammetry was not used here due to the enormous time constants involved.

Discussion

The obtained results can be attributed to the various aspects of the device. A possible explanation is that a stainless steel electrode is a combination of different elements (nickel, chromium, manganese and iron), hence, not chemically inert. Therefore, like any other non-inert metal, it will form an insulating layer interface between the metal and the polymer.^6,19 It is covered by a very thin layer (a few nanometers) of Cr₂O₃ inhibiting further oxidation of the steel. Cr₂O₃ is an electric insulating material. If made as a very thin layer (a few nanometers) it creates a metal (electrode): insulator interface in the device. Electric conduction becomes possible through special mechanisms like the Schottky effect which brings change in electrostatics between a metal/semiconductor interface or through field emission.²⁰

It was mentioned recently in literature that the diffusion of silver ions into PEDOT:PSS would be responsible for the observed phenomena for the silver coated polyamide yarns.^12,21,22 Our results obtained with stainless steel electrode yarns, however, prove that this cannot be the only conduction mechanism in the PEDOT:PSS device.

A possible mechanism of charge storage could be that PEDOT:PSS is a dipole, and initially the molecules are randomly arranged. On application of an electric field, polarization takes place, the cations PEDOT+ are attracted to the negative electrode while the anions PSS- are attracted to the positive electrode.

The positive PEDOT and negative PSS ions being quite large as indicated in Figure 8 explain the long time needed to disorganize them. The charging brings in a dielectric polarization of PEDOT:PSS, and on release, relaxation takes place as the molecules try to get back to a random orientation. Some differences in the performance of the silver coated yarn electrodes compared to the stainless steel filament yarns could be due to the difference in the ohmic contact between the polymer and the different electrode.²³ It must be emphasized here that the physics of PEDOT:PSS is still under debate as claimed by other authors.^14,15
Figure 8.
Chemical structure of PEDOT:PSS.

Conclusion

Textile-based charge storage devices with PEDOT:PSS have been fabricated. In one set of devices, pure stainless steel filament yarns were used as electrodes; in a second set, silver coated PBO yarns were applied. From the results, the pure stainless steel filament yarn electrode performed better than the silver coated yarn electrode device for the fact that it can store its charge for a longer period of time and at a higher value. This energy storage device has been well integrated into a textile structure making it light weight and flexible; hence, it fits well with wearable garments. The device can be easily fabricated. The results motivate the making of a functional textile battery integrated within the textile structure, with improved performance, reliability and safety in further research.

Footnotes

Funding

Sheilla A Odhiambo, on study leave from Moi University, Eldoret, Kenya, wants to thank the coordinators of the MU-VLIR UOS project for the financial support for her research and stay at Ghent University.

References

Muller

Hamedi

Karlsson

. Woven electrochemical transistors on silk fibers. Adv Mat 2011; 23: 898–901.

Carpi

De Rossi

. Electroactive polymer-based devices for e-textiles in biomedicine. IEEE Trans Info Tech Biomed 2005; 9: 574–574.

Service

. Technology – electronic textiles charge ahead. Science 2003; 301: 909–911.

Horng

Y-Y

Y-C

Hsu

Y-K

. Flexible supercapacitor based on polyaniline nanowires/carbon cloth with both high gravimetric and area-normalized capacitance. Power Sources 2010; 195: 4418–4422.

Van Langenhove

Hertleer

. Smart clothing: a new life. Inter Clothing Sci Tech 2004; 16: 63–72.

Sundfors

Gustafsson

Ivaska

. Characterisation of the aluminium-electropolymerised poly(3,4-ethylenedioxythiophene) system. Solid State Electrochem 2010; 14: 1185–1195.

Catrysse

Puers

Hertleer

. Towards the integration of textile sensors in a wireless monitoring suit. Sens Act A-Physl 2004; 114: 302–311.

Salkind AJ. Advances in battery technologies and markets: Material science aspects. In: Ginley DS, Doughty DH, Scrosati B, et al. (eds) Materials for electrochemical energy storage and conversion Ii-batteries, capacitors and fuel cells. Cambridge University Press: Cambridge, 1998, pp. 3–14.

Hu L, La Mantia F, Hui W, et al. Lithium-ion textile batteries with large areal mass loading. Advanced energy material, pp. 1012–1017, http://onlinelibrary.wiley.com/doi/10.1002/aenm.201100261/pdf (2011, accessed 25 November 2011).

10.

Kaltenbrunner

Kettlgruber

Siket

. Arrays of ultracompliant electrochemical dry gel cells for stretchable electronics. Adv Mat 2010; 22: 2065–2067.

11.

Liu

Gorgutsa

Santato

. Flexible, solid electrolyte-based lithium battery composed of LiFePO4 cathode and Li4Ti5O10 anode for applications in smart textiles. Electrochem Soc 2012; 159: A349–A356.

12.

Bhattacharya

De Kok

Zhou

. Rechargeable electronic textile battery. Appl Phys Let Article No. 223305/3. 2009; 95.

13.

Hernandez-Labrado

Contreras-Donayre

Collazos-Castro

. Subdiffusion behavior in poly(3,4-ethylenedioxythiophene): polystyrene sulfonate (PEDOT:PSS) evidenced by electrochemical impedance spectroscopy. Electroanal Chem 2011; 659: 201–204.

14.

Ouyang

Chu

. On the mechanism of conductivity enhancement in poly (3,4-ethylenedioxythiophene): poly(styrene sulfonate) film through solvent treatment. Polymer 2004; 45: 8443–8450.

15.

Lee

Park

. Electrochemistry of conductive polymers. 45. Nanoscale conductivity of PEDOT and PEDOT:PSS composite films studied by current-sensing AFM. Phys Chem B 2010; 114: 2660–2666.

16.

Rattfalt

Linden

Hult

. Electrical characteristics of conductive yarns and textile electrodes for medical applications. Med Bio Engr Comp 2007; 45: 1251–1257.

17.

Kayacan

Bulgun

Sahin

. Implementation of steel-based fabric panels in a heated garment design. Text Res 2009; 79: 1427–1437.

18.

Schwarz

Kazani

Cuny

. Comparative study on the mechanical properties of elastic, electro-conductive hybrid yarns and their input materials. Text Res 2011; 81: 1713–1723.

19.

Inganas

Lundstrom

. Electronic properties of metal polypyrrole junctions. Synth Met 1984; 10: 5–12.

20.

Solymar

Walsh

. Electrical properties of materials, 6th ed. Oxford: Oxford University Press, 1997, pp. 386–386.

21.

Mousavi

Bobacka

Ivaska

. Potentiometric Ag+ sensors based on conducting polymers: a comparison between poly(3,4-ethylenedioxythiophene) and polypyrrole doped with sulfonated calixarenes. Electroanal Chem 2005; 17: 1609–1615.

22.

Vazquez

Bobacka

Luostarinen

. Potentiometric sensors based on poly(3,4-ethylenedioxythiophene) (PEDOT) doped with sulfonated calix 4 arene and calix 4 resorcarenes. Solid State Electrochem 2005; 9: 312–319.

23.

Shen

Hosseini

Wong

. How to make ohmic contacts to organic semiconductors. Chem Phys Chem 2004; 5: 16–25.