Abstract
In our solar textiles constructed from the direct integration of photovoltaic (solar) cells onto woven polyester fabric, the fabric has to be rendered electrically conducting.
Conductivity is conferred on the fabric by the deposition of thin layers of first a conducting polymer and then aluminium. This approach seeks to circumvent serious disruptions to conductivity caused by any cracking of the aluminium layer during installation or use of the solar textile. In addition, the thermal stability of the treated fabric also needs to be assured during deposition of the photovoltaic cells, a process conducted over 20 minutes at 200°C.
It is clear, therefore, that the viability of our strategy needs to be thoroughly tested, and in this paper we discuss the electrical, thermal and mechanical tests that have been carried out to date. Two conducting polymers, polyaniline and poly(3,4-ethylenedioxythiophene) poly(styrenesulfonate) (PEDOT:PSS), have been included in our tests. The results from fabrics containing PEDOT:PSS have been particularly encouraging in terms of the durability of our conducting polyester fabric.
Key words: solar, textiles, polyaniline, PEDOT:PSS, conductive, tests
Introduction
In recent years we have been exploring the deposition of silicon (photovoltaic) cells directly onto standard polyester fabrics. This quest has been instigated through the realisation that amorphous/nanocrystalline films of silicon can be successfully deposited at a temperature as low as 200°C (Shirai et al, 1999, Löffler et al, 2006), a temperature well below the melt temperature of polyester fabrics (250-260°C), whereas the deposition of most other types of solar cell requires temperatures of at least 400°C.
The configuration of our solar textile cell bears close resemblance to that of a conventional solar cell that utilises a glass or polycarbonate base (Fig. 1). As in a conventional solar cell, our solar textile includes two dissimilar doped semiconducting layers, whereby an electric field is built into the cell, to produce a so-celled PN junction.
When light falls onto the cell, the junction separates the negative electrons generated, from the positive ‘holes’ that are left behind. The resulting electrical potential is typically 0.5 V or slightly higher.
In our solar textile construction, as with a number of conventional constructions, the doped P and N layers are separated by an undoped intrinsic I layer. On top of the P layer is a thin transparent, electrically conducting layer, usually formed of indium tin oxide (ITO), though we are exploring the use of a modified zinc oxide as an alternative. The fabric can then be protected by a suitable encapsulant.
The textile fabric, which forms the substrate for the solar cell, also has to be rendered electrically conducting. In conventional solar cells, the glass or polycarbonate base is rendered electrically conducting by treatment with a metal layer, often aluminium. In our solar textile construction, conductivity is conferred on the fabric through the deposition of thin layers of first a conducting polymer and then one of aluminium.
This approach is aimed at circumventing serious disruptions to conductivity caused by any cracking of the aluminium layer during installation or use of the solar textile. Moreover, the thermal stability of the treated fabric also needs to be assured during deposition of the photovoltaic cells, a process conducted over 20-30 minutes at 200°C. The viability of our approach has therefore required thorough testing, and in this paper we discuss the electrical, thermal and bending tests that we have carried out to date.
Experimental
The polyester fabric samples, supplied by J&D Wilkie Limited, were plain weave, with 210 filaments per inch in the warp and weft directions and a total loomstate weight of 250 g m-2. Two commonly applied commercial conductive polymers were used: polyaniline (Panipol W, 6 to 10 wt% in water) supplied by Panipol Oy, Finland and poly(3,4-ethylenedioxythiophene) –poly(styrenesulfonate), known as PEDOT:PSS, supplied by Heraeus Clevios GmbH, Germany. The fabric samples were cleaned in 1% Decon detergent in deionised water for 30 min in an ultrasonic bath, before being rinsed in running deionised water for 15 min and dried in a flow of hot air. After calendering for 1 min at 240°C, a low pressure air plasma treatment was performed on the fabric. A number of layers of conducting polymer were then applied to the fabric. Finally, a layer of ca. 100 nm of aluminium was evaporated onto the polymer coated fabric. A more detailed description will be provided elsewhere (Diyaf et al, 2013).
Electrical resistance of each fabric was measured using the standard methods, BS6524 and AATCC Test Method 76, 84-2005 (BS 6524, 1984; 76-2005 A.t.m., 2010; ASTM, 1999). Thermal annealing measurements were conducted, in order to study the effects of temperature changes in vacuo up to 200°C on the electrical properties of each of the two conductive polymers. In terms of electrical properties, the solar fabrics must withstand the silicon deposition process, which occurs at 200°C under a pressure of 0.2 torr, using a low power electrical plasma to decompose gaseous precursors. In each thermal annealing test, a sample was heated from room temperature up to 200°C, and simultaneously its resistance was measured continuously. When the temperature reached 200°C, the sample was held at this temperature for 40 min, before the heater was switched off. Resistance and temperature were then both continuously recorded, as the sample cooled to room temperature.
Two types of bending test were conducted. In the first type, the sample was bent vigorously for 100 s, and its surface resistivity was continuously monitored. After vigorous bending, the surface resistivity was then monitored for a further 100 s. In some cases, this cycle was repeated. In the other test, the sample was bent systematically over a range of angles, at which values of its resistivity were determined. More detailed descriptions of all these tests will be given elsewhere (Diyaf et al, 2013).
Results and Discussion
Surface Resistivity
Fig. 2 shows the reduction in surface resistivity with increasing number of layers of polyaniline. After six or seven layers have been applied, the resistivity appears to remain constant, having fallen by ca. 85% from that determined where there is only one layer. However, a large number of layers cannot readily be applied in practice, because of the high viscosity of the polyaniline solution. Fig. 2 also shows the reduction in surface resistivity with successive application of layers of PEDOT:PSS. The surface resistivity is reduced by ca. 95% to a constant value after only four layers have been applied. Indeed, the application of just three layers leads to a fall in surface resistivity of 90%.
Thermal Annealing Tests
Fig. 3 shows changes in surface resistivity that fabrics coated with polyaniline exhibited during thermal cycling to 200°C and back to room temperature. During heating, the surface resistivity hardly rose at all. However, while the temperature was maintained at 200°C, slight rises in surface resistivity were observed. After the heater was switched off, surface resistivity increased dramatically, particularly for the fabric with only one layer of polyaniline. The increase in surface resistivity could be attributed to thermally induced morphological and chemical changes, as discussed elsewhere (Chandrakanthi and Careem, 2000).
The corresponding results for a fabric coated with three layers of PEDOT:PSS are shown in Fig. 4. During heating and also while the temperature was maintained at 200°C, large fluctuations in surface resistivity were observed. However, after the heater had been switched off and the temperature had fallen below 90°C, the surface resistivity returned to its original value. The heating cycle appears not have affected the nature of the PEDOT:PSS layers. Thus, PEDOT:PSS films on our polyester fabric seem more stable than polyaniline films.
Fig. 5 shows changes in surface resistivity for polyaniline coated samples onto which a layer of aluminium had been evaporated. As expected, the surface resistivity values were considerably reduced by the presence of the aluminium layer. While the temperature was kept constant at 200°C, fluctuations in surface resistivity were observed, particularly for the fabric with two layers of polyaniline. However, on cooling to room temperature, surface resistivities returned to their initial values.
Fig. 6 shows corresponding results where the fabric has been coated with three layers of PEDOT:PSS. Again, fluctuations in surface resistivity were observed while the temperature was maintained at 200°C, but on cooling the surface resistivity fell to a constant value slightly below the initial value. Reports have been published that heating a PEDOT:PSS film to 200°C activates charge carriers and increases crystallinity in the film, thereby resulting in improved conductivity (Huang et al, 2003).
Long-term Stability Tests
Fig. 7 shows changes in surface resistivity of fabrics coated with either polyaniline or PEDOT:PSS. The surface resistivities of both fabrics were observed to decrease over the first five hours of measurement. The surface resistivity of the fabric coated with PEDOT:PSS remained constant, whilst that of the fabric coated with polyaniline gradually climbed back to its starting value. These results also provide evidence for the greater stability of PEDOT:PSS films over polyaniline films on our polyester fabric.
Bending Tests
Fig. 8 shows bending test results for fabric samples coated with layers of conducting polymer and one layer of aluminium. In the sample containing two layers of polyaniline, a massive increase in resistance was observed, which we attribute to the presence of micro-cracks and other microstructural damage observed with the optical microscope.
Fig. 8 also shows consecutive bending tests for a sample containing three layers of PEDOT:PSS. In this case, the conductivity of the fabric was maintained, and indeed after the first test, the conductivity actually increased. These results indicate that the PEDOT:PSS coating is quite stable, when the fabric is roughly handled.
Systematic Tests
Fig. 9 shows changes in resistance with bending angle, as each sample is first progressively bent from 180° (no bending) down to 50°, and then in a reverse stage back to 180°. In the case of the sample containing two layers of polyaniline, only slight increases in resistance are observed down to an angle of 110°. Below 110°, the resistance increases markedly down to 50°. In the reverse stage, the changes in resistance appear more erratic, though it can be noted that the resistances are much higher than the equivalent values measured in the first stage. This increase in resistance can again be attributed to the onset of micro-cracks in the conducting layers, as a result of bending the fabric.
The changes in resistance observed for the sample containing three layers of PEDOT:PSS appear much smoother. In the first stage, no significant change in resistance is observed until the fabric is bent below 90°. In the reverse stage, the resistance progressively falls to almost its initial value at an angle of 110°, whereupon it stays effectively constant. Again, it can be noted that the PEDOT:PSS coating appears to be the more durable one.
Conclusion
Our results demonstrate the viability of our approach to rendering polyester fabrics electrically conducting and to maintaining conductivity under harsh thermal and mechanical treatments. Thus, the tests indicate the types of deformation that can be sustained in applications of our solar fabrics. It is concluded that for these purposes, layers of PEDOT:PSS are more durable than those of polyaniline.
Both polymers are well established for their electrically conducting properties. Conduction is considered to depend primarily on the morphology and structure of the polymers. Thus, electrically conducting polyaniline has been depicted as consisting of conducting crystalline regions separated by more poorly conducting amorphous regions (Pelster et al, 1994). A similar picture has evolved for PEDOT:PPS (Huang et al, 2003). We suggest too that the texture of the woven polyester fabric base can influence the nature of the two regions, and their relative proportion. In the case of polyaniline, heat tends to reduce the size of the conducting regions, in parallel with broadening of the space between them, so that overall electrical conductivity decreases. The temperature at which this decrease in conductivity becomes manifest depends on the nature of the polyaniline (Pelster et al, 1994). However, in the case of PEDOT:PSS, there is evidence that heating can coalesce constituent particles, such that a more crystalline structure is adopted (Huang et al, 2003).
Conductivity is then enhanced. Too high a temperature will, though, cause some degradation of PEDOT:PSS, and so reduce conductivity. It appears, therefore, that in our tests, notably the long term stability tests and the thermal annealing tests, the polymer chains in the grade of polyaniline used suffer some degradation, whilst those comprising the PEDOT:PSS films remain stable.
Acknowledgements
John Andrews, Suzanne Jardine and Helena Lind are thanked for their contributions towards the development of the concept of our solar textile.
We are also grateful for financial support from: the Scottish Government (SMART:SCOTLAND funding), Scottish Enterprise, the Engineering and Physical Sciences Research Council, the Scottish Optoelectronics Association (TTOM funding), the Scottish Universities Physics Alliance and RADIKAL.
Samples of woven polyester from J&D Wilkie Limited, Kirriemuir, are gratefully acknowledged. We thank Emiliano Rezende for a sample of PEDOT:PSS.
References
76-2005 A.t.m. (2010), Electrical resistivity of fabrics, AATCC Technical Manual.
ASTM (1999), Standard D257-99. Standard test methods for DC resistance or conductance of insulating materials.
BS6524 (1984), British standard method for determination of surface resistivity of a textile fabric.
Chandrakanthi N & Careem M A (2000), Thermal stability of polyaniline, Polymer Bulletin, 44, 101-108.
Diyaf A, Wilson J I B & Mather R R (2013), paper in preparation.
Huang, J, Miller, P F, de Mello J C, de Mello A J & Bradley D D C (2003), Influence of thermal treatment on the conductivity and morphology of PEDOT/PSS films, Synthetic Metals, 139, 569-572.
Loffler J, Devilee C, Geusebroek M, Soppe W J & Muffler H-J (2006), Deposition of ƒÝc Si:H by microwave PECVD ¡V influence of process conditions on layer properties, in Proceedings of the 21st European Photovoltaic Solar Energy Conference, Dresden, Germany, p1596-1599.
Pelster R, Nimtz G & Wessling B (1994), Fully protonated polyaniline: Hopping transport on a mesoscopic scale, Physical Review B, 49, 12718-12723.
Shirai H, Sakuma Y, Moriya Y, Fukai C & Ueyama H (1999), Fast Deposition of Microcrystalline Silicon Using High-Density SiH4 Microwave Plasma, Japanese Journal of Applied Physics 38, 6629-6635.