Waterproof and durable screen printed silver conductive tracks on textiles

Abstract

Conductive textiles are fabrics that include conductive yarns woven into or conductive tracks printed on to the textiles. Conductive textiles have attracted significant attention, since they are fundamental for the integration of electronic functions to achieve wearable devices. Screen printing is a well-established and cost-effective fabrication method; it enables a versatile layout of conductive tracks. The limitation of the current screen-printed conductive textiles is low durability to weathering, abrasion and washing. This paper presents a process for producing a waterproof and durable conductive textile using only screen printing. A three functional layer design was used to fabricate the durable conductive tracks. Firstly, an interface layer was printed to provide a smooth surface for subsequent printing, under-side protection and electrical insulation. Next, a silver layer provided the conductive track and finally an encapsulation layer was printed on top to provide upper-side protection and electrical insulation. The printed silver tracks achieved maximum conductivity using a single print. The conductivity of the silver tracks returned to its original value when they were dried after soaking in water continuously for 24 hours.

Keywords

waterproof conductive fabric interface layer encapsulation layer printed electronics smart fabric wearable electronics

Wearable electronics have been used in various applications, such as patient monitoring, fabric based electronic entertainment devices and electronic functional uniforms for the military.^1–3 Wearable electronic systems typically require conductive tracks to form electrical interconnections between components. These conductive tracks can be realized using two techniques as follows.

Combining conductive yarns into the textile by weaving, knitting or embroidery.^4–6 The disadvantages of weaving and knitting are the limitation on design freedom because the conductive path is constrained to follow the physical location of the yarns within the fabric. Embroidery offers higher flexibility of design but it is an expensive fabrication method especially for achieving conductive geometries with fine resolutions. It requires high quality metal-coated polymer yarns (e.g. silver coated nylon), since metal fiber yarns are not suitable for machine stitching due to the low yarn elasticity and build-up of tension during stitching.⁷

Printing conductive paste/ink on to textiles.^8–14 Printing is a straightforward and versatile fabrication method that gives complete freedom for the design layout, since it does not need to follow the fabric structure. The challenge for printed electronics on fabrics is durability and, in particular, withstanding bending, stretching, abrasion and washing.

To improve the durability of screen-printed conductive textiles, coating and lamination methods to protect the conductive elements have been investigated by many researchers. Karaguzel et al.¹⁰ investigated the lamination of a thermoplastic polyester (PES)-based urethane elastomer to protect the screen printed conductive silver on two non-woven fabrics. Kim et al.^11,12 investigated the washing and bending resistance of conductive PES woven fabric on which silver conductive layers were screen printed. The conductive layers were successfully passivated but the exact coating material or method was not specified. Kazani et al.^13,14 improved the washing durability of screen-printed silver on woven fabrics and the dry cleaning durability of screen printed silver on both a polyurethane flexible foam and a non-woven PES by laminating a thermoplastic polyurethane layer over the conductive silver. This research demonstrated that the additional protective layers provide improved durability, but in each case, whilst the conductive tracks were screen printed, the protective layers used another deposition process (lamination or coating).^10–14

In this work, we report the evaluation of conductive textiles with protective layers fabricated entirely by screen printing. The use of screen printing offers several advantages over lamination and coating methods. Factors include design flexibility, which is particularly important for realizing complicated designs. Less material is wasted because the protective material is only printed where needed over the conductive track. This also reduces the impact on the inherent properties of the textile (e.g. flexibility, lightweight, breathability), which is further reduced by printing the finest patterns and thinnest films possible. In this work, we report an investigation of ultraviolet (UV)-curable polyurethane acrylate pastes (PUs) as the protective layers. UV curing is a rapid technique performed at room temperature and is a more environmentally friendly fabrication method, when compared to thermal curing, due to the low release of volatile organic compounds (VOCs). Polyurethane is a flexible material with excellent adhesion to textiles.¹⁵

Two protective layers were used to seal the conductive track. First an interface layer was printed directly onto the fabric. The conductive track was then printed on the interface layer. After printing the conductive tracks, the protective encapsulation layer was then printed, overlapping the interface layer to seal in the conductive layer in a sandwich structure. Two screen-printable UV curing pastes have been investigated for these layers: Fabink-UV-IF1 (general purpose PU) and Fabink-UV-IF010 (waterproof PU).¹⁶ These materials have been used individually or combined to improve the waterproofing and adhesion properties. The combinations are shown in exploded view and cross-section through the track in Figure 1.

Figure 1.

Exploded and cross-section view of the completed design using single protective material (a) and dual protective materials (b).

Materials and screen printing

Materials

Fabric

PES/cotton (65/35) was chosen for the fabric because it has widespread use in the garment industries and gives an example of a typical fabric that could be used in smart fabric applications. The fabric was provided by Klopman International and its properties are shown in Table 1.

Table 1.

Fabric properties

Fiber content	Yarn density (threads/cm)	Weave type	Weight (g/m²)	Thickness (mm)
Polyester 65% Cotton 35%	Warp 48 Weft 23	2 × 1 twill	210	0.316

Interface and encapsulation pastes

Two UV-curable PUs, Fabink-UV-IF1 (general purpose interface paste) and Fabink-UV-IF010 (waterproof interface paste), were obtained from Smart Fabric Inks Ltd, UK. These pastes were chosen because they are compatible with a wide range of fabrics and they provide good adhesion (Fabink-UV-IF1) to other pastes and waterproofing property (Fabink-UV-IF-010). The paste properties are shown in Table 2.

Table 2.

Ultraviolet (UV)-curable polyurethane acrylate paste properties

Paste name	Fabink-UV-IF1	Fabink-UV-IF010
Solids	100%	100%
Viscosity	4.7 Pa.s	10.5 Pa.s
Cured thickness with 250 mesh	40 µm	35 µm
Curing time	30 s	60 s

Conductive silver

Table 3 shows the properties of Fabink-TC-AG1 silver paste used in this study, which was obtained from Smart Fabric Inks Ltd. It is a screen-printable paste that mainly consists of silver flake, vinyl resin and solvent. Fabink-TC-AG1 was chosen in this study because it has a low curing temperature and is compatible with the Fabink-UV-IF1 interface material. The silver track has good conductivity and is flexible after curing.

Table 3.

Conductive silver paste properties

Properties	Fabink-TC-AG1
Viscosity	3.5 Pa.s
Binder	Vinyl resin based
Solids	58–62%
Sheet resistivity	30 mΩ/□
Cure condition	10 minutes at 120℃

Fabrication overview

The screen-printing process and the materials used in each printing step are shown in Figure 2. For the single material structure, as shown in Figure 1(a), the interface layer was formed using four prints of the paste (Fabink-UV-IF1 or Fabink-UV-IF010). For the dual material structure, as shown in Figure 1(b), the interface was formed using three prints of Fabink-UV-IF010 followed by one print of Fabink-UV-IF1 using a 250 thread/inch stainless steel screen with UV curing, which was applied after each print. The conductive layer was achieved with a single print of Fabink-TC-AG1 using a 120 thread/cm PES screen and it was cured at 120℃ for 10 minutes in a box oven. For the single material structure, the encapsulation layer was formed by four prints of the same paste used as the interface layer. The dual material structure used one print of Fabink-UV-IF1 followed by three prints of Fabink-UV-IF010 and again UV curing was applied after each print. An UV 500 W curing cabinet (UV Light Technology Ltd) with an iron (Fe) bulb, which can provide a wavelength from 160 nm upwards, was used in this study.

Figure 2.

Fabrication processing and the materials used in each printed layer.

Interface layer printing

In contrast to most printed electronic textiles, which deposit the conductive material directly onto the textile, this work used a screen-printed interface layer. This approach produces a smooth and flexible surface layer on to which the silver paste can be printed. This ensures that the silver layer provides low sheet resistance (80 mΩ/□) with a single print of silver paste on the interface layer.¹⁷ This avoids the need to print numerous silver layers, which is a comparatively expensive material. The interface can be applied to smooth the surface of a variety of textile types, making this approach useful in a wide range of applications.

A semi-automatic standard screen printer, DEK248, shown in Figure 3, was used in this work. Four prints with a total thickness of ∼120 µm were required to obtain a smooth interface.

Figure 3.

DEK248 semi-automatic screen printer.

Conductive layer printing

Fabink-TC-AG1 silver paste was printed on top of the interface layer to produce 30 mm long, 1 mm wide conductive tracks with 2 mm × 2 mm contact pads at each end. The thickness of the silver layer is 5–7 µm. The resistance of the conductive tracks was measured using a FLUKE 115 Multimeter.

Encapsulation layer printing

For the purpose of this study, before printing the encapsulation layers, the conductive contact pads were masked off to prevent the encapsulation pastes depositing directly onto the connection pads and preventing future measurement. The masking material was removed when required to reveal the conductive silver contact pads. The exposed pads were used to measure resistance in subsequent experiments. The thickness of the encapsulation layers with four prints was ∼90 µm. Visual inspection showed this thickness provided good step coverage over the conductive tracks.

Test methods

Water absorption of the PU film

In order to evaluate the waterproofing property of the protective materials, the degree of water absorption by the PU films was measured using 1 mm thick samples of just the cured PU pastes prepared by casting in a poly(tetrafluoroethylene) mold. The UV curing time was 60 s for Fabink-UV-IF1 and 120 s for Fabink-UV-IF010 to completely cure 1 mm thick samples. The films were trimmed to 60 mm × 15 mm and weighed prior to immersion in deionized water. After 24 hours, the samples were taken out and weighed again after wiping off the surface water. The percentage water absorption was calculated by Equation (1):

W (%) = \frac{w_{t} - w_{0}}{w_{0}} \times 100 %

(1)

where w₀ and w_t are the weight before and after 24 hours immersion in the water.

The percentage length change due to the swelling of the samples after water absorption was also calculated, using Equation (2):

l (%) = \frac{l_{t} - l_{0}}{l_{0}} \times 100 %

(2)

where l₀ and l_t are the length of the samples before and after 24 hours immersion in the water.

Contact angle and surface energy

The surface energy of the interface layer affects the adhesion between the interface layer and the subsequent silver layer. There is no direct method to measure the solid surface energy. The contact angles (θ) shown in Figure 4 of at least two liquids with known surface tension in terms of its dispersive and polar components are required to calculate the surface energy. These two liquids are used as probe liquids on the surface of the solid of which the surface energy is to be calculated.

Figure 4.

Liquid contact angle on the solid sample.

In order to obtain the surface energy for the interface materials (Fabink-UV-IF1 Fabink-UV-IF010), the contact angles of water and diiodomethane were measured using a Krűss DSA30B Tensiometer.

The total surface tension γ and its dispersive part

γ^{D}

and polar part

γ^{P}

are given in Table 4. Measurements were taken from at least five different positions on each film type and the average values for the contact angles were calculated.

Table 4.

Surface tension of water and diiodomethane

Probe liquid	Surface tension (mN/m)
Probe liquid	γ ^D	γ ^P	γ
Water	21.8	51	72.8
Diiodomethane	50.8	0	50.8

The extended Fowkes method shown in Equation (3) was used to calculate the surface energy.¹⁸ This method is applicable to determine the surface energy of any solid:^18,19

r_{L} (1 + \cos θ) = 2 (\sqrt{r_{L}^{D} r_{S}^{D}} + \sqrt{r_{L}^{P} r_{S}^{P}})

(3)

r_{S} = r_{S}^{D} + r_{S}^{P}

(4)

where

γ_{L}

is the surface tension of the probe liquid and

γ_{L}^{D}

and

γ_{L}^{P}

are the dispersive and polar components, respectively. θ is the contact angle of the probe liquid.

γ_{S}

is the surface energy of the solid film and

γ_{S}^{D}

and

γ_{S}^{P}

are the dispersive and polar components of the solid film, respectively. By measuring the contact angles of water and diiodomethane, we can resolve the Equation (3) with two unknown (

γ_{S}^{D}

and

γ_{S}^{P}

.) and therefore obtain the surface energy

γ_{S}

of the solid film in Equation (4).

The change of water contact angle on both interface materials in 120 s was also recorded using the live camera fitted in the equipment. This was done to investigate the hydrophilic/hydrophobic properties of the interface materials.

Bending tests

To evaluate the flexibility of the conductive fabric using the single and dual protective material approaches, the resistance of the silver track after bending was measured. This was achieved by bending the conductive tracks around a cylindrical rod with a diameter of 10 mm.

Scanning electron microscope (SEM) of the fabric and fabric with different interface layers

SEM micrographs obtained using a Zeiss EVO LS25 were used to examine the surface roughness of the fabric and the effectiveness of different thicknesses of the interface layer.

Waterproof test

The waterproofing properties of the protective layers were investigated by immersing the conductive fabric samples in deionized water. The samples were removed after continuous immersion for 24 hours and dried at 40℃ for 30 minutes in a box oven then left to cool down in air before measuring the resistance of the conductive tracks. At least five samples were used for each test to obtain a representative average value.

Results and discussion

Water absorption of the interface materials

The water absorption and length change of the two PU samples after immersion in water for 24 hours are shown in Table 5. The water absorption by weight of the Fabink-UV-IF1 film is 18.9%, with the length increasing by 6.7%. The waterproof polyurethane, Fabink-UV-IF010, absorbed less than 0.5% water in 24 hours and there was no measurable change in length.

Table 5.

Water absorption and length change of the interface films in 24 hours

Sample	Water absorption	Length change
Fabink-UV-IF1	18.9% ± 0.5%	6.7% ± 0.1%
Fabink-UV-IF010	0.3% ± 0.1%	0

Most polyurethane films have poor water resistance because of the hydrophilic groups, such as the carboxyl group, in the molecular structure. The dimensions of the film increase after absorbing water and the consequent increase in the length/width of the interface/encapsulation films stretch the silver, causing an increase in the resistance of the silver tracks and potentially breaking them. The results show that Fabink-UV-IF010 is an effective waterproof bottom interface layer and top encapsulation layer material that prevents water penetration into the conductive tracks and avoids dimensional changes.

Surface energy of the protective films and water contact angles on the films

Table 6 shows the contact angle and surface energy of Fabink-UV-IF1 and Fabink-UV-IF010 films. The Fabink-UV-IF010 film has a lower surface energy than the Fabink-UV-IF1 film, which affected the adhesion of the silver film, as discussed in the next section.

Table 6.

Contact angle and surface energy of two polyurethane acrylate paste films

Film	Contact angle (°)		Surface energy (mN/m)
Film	Water	Diiodomethane	$γ_{S}^{D}$	$γ_{S}^{P}$	$γ_{S}$
Fabink- UV-IF1	81.8 ± 1.2	47.8 ± 0.7	35.4 ± 0.4	3.7 ± 0.4	39.2 ± 0.8
Fabink- UV-IF010	105.5 ± 0.2	57.0 ± 0.5	30.3 ± 0.3	0	30.3 ± 0.3

To examine the wettability of the two PU films, water drop images were recorded for 120 s. Figure 5 shows the water contact angle changes on two PU films as a function of time. The water contact angle on Fabink-UV-IF1 film decreased from 81.8° to 44.4° in 120 s. The water contact angle on Fabink-UV-IF010 film decreased from 105.5° to 92° in the same time. The results show that Fabink-UV-IF1 film is hydrophilic, whereas Fabink-UV-IF010 is highly hydrophobic.

Figure 5.

Water contact angle changes in 120 seconds on the two polyurethane acrylate paste films.

Flexibility of the conductive tracks on different interface layers

Figure 6 shows the flat conductive silver tracks with different protective materials. Figure 7 shows the bending test applied to the conductive tracks using a cylindrical rod with a diameter of 10 mm.

Figure 6.

Conductive tracks before bending: (a) silver with Fabink-UV-IF1; (b) silver with Fabink-UV-IF010; (c) silver with Fabink-UV-IF1 and Fabink-UV-IF010.

Figure 7.

Conductive tracks bent around a cylindrical rod with a 10-mm diameter: (a) silver track with Fabink-UV-IF1; (b) silver track with Fabink-UV-IF010; (c) silver track with Fabink-UV-IF1 and Fabink-UV-IF010.

The fabric showed good flexibility after the printing of the interface, conductive and encapsulation layers. There were no cracks on the conductive tracks without bending, as shown in Figure 6. Under the bending condition, there were no obvious cracks on the conductive tracks with the Fabink-UV-IF1 single material protective layers (Figure 7(a)) and Fabink-UV-IF1 and Fabink-UV-IF010 dual material protective layers (Figure 7(c)). However, the conductive track with the Fabink-UV-IF010 single material protective layers broke (Figure 7(b)) and therefore had no conductivity. It is therefore not practical to print Fabink-TC-AG1 silver paste directly on the waterproof interface layer, as the adhesion between the silver and interface layer was poor and the silver layer would be expected to crack on bending.

Scanning electron microscope analysis

The scanning electron microscope (SEM) images in Figure 8 show both the surface and the cross-section of the bare fabric and with different thickness interface layers.

Figure 8.

Scanning electron microscope images of the fabric (a), fabric with two prints (b) and four prints (c) of the interface layer.

The fabric has a rough surface, as shown in Figure 8(a), and printing a conductive silver layer directly onto this surface would require more silver paste to bridge the gaps between the yarns. This wastes material and is expensive. In addition, the silver layer would be less homogenous along its length and this would affect the repeatability of the printed tracks and its conductivity. The printed interface layer on top of the fabric reduces the surface roughness of the fabric, thus saving expensive silver paste. As the interface layer increases, the defects on the surface are reduced and the surface becomes smoother. For the PES cotton used in this study, four prints were required to obtain a smooth interface layer, as shown in Figure 8(c). The number of prints needed for the interface layer to achieve a smooth surface is determined by the characteristics of the fabric. Smooth fabrics with fewer loose fibers and a tighter weave need fewer prints than rough and loose fabrics. Non-woven fabrics typically have a smoother surface than woven fabrics, which may only need a single interface layer to achieve a smooth surface. For woven fabric, the yarn density, weaving method and fabric thickness can also affect the surface roughness and the gap between the yarns, thus affecting the number of prints in the interface layer.

The interface layer also provides an all round protective layer for the conductive silver, as damage may occur from underneath the fabric, for example water penetration to the silver through the gaps between the yarns. Encapsulation of the conductive tracks printed on the fabric from the top therefore is not sufficient to fully protect the conductive silver layer. The combination of interface layer and encapsulation layer improves the durability of the conductive tracks that are sandwiched between the two protective layers.

Waterproof test of the printed conductive silver

Figure 9 shows the resistance changes of the single protective material and dual protective material designs. For the Fabink-UV-IF1 single protective material design, the resistance of the silver tracks increased due to absorption of the water, which causes the interface/encapsulation layers to expand. Also, the silver tracks were bent due to the bending of the interface/encapsulation layers after immersion into the water. The stretching and bending strains the silver track, causing the resistance to increase by 30%.

Figure 9.

Resistance changes after immersion into water for 24 hours.

For the Fabink-UV-IF010 single protective material design, the resistance increased by 8%. Fabink-UV-IF010 is hydrophobic and so the water absorption is very low (<0.5%). The dimensions of the Fabink-UV-IF010 interface/encapsulation did not change but the samples were slightly bent after immersion into water for 24 hours, due to the water absorption of the fabric. The bending strains the silver tracks, although the samples became flat again after drying.

For the dual protective material design, the resistance stayed the same when the samples were dried after immersion into water for 24 hours. The Fabink-UV-IF010 bottom interface layer and top encapsulation layer prevents the absorption of water, while the Fabink-UV-IF1 top interface layer and the bottom encapsulation layer provides a good adhesion between the silver and the interface/encapsulation layers. The dual protective material design effectively increased the resistance to water of the conductive fabric while maintaining good flexibility.

Conclusions

The three-layer design is an effective method to produce conductive tracks on fabric. The interface layer provides a smooth surface on fabric, therefore obtaining good conductivity using a minimum of silver paste. It also serves as a protective layer, together with the encapsulation layer, to improve the durability of the conductive track.

The dual protective material design combines the advantage of the waterproof property of the hydrophobic PU and the good adhesion property of the hydrophilic PU. The two PUs have good adhesion to each other due to the strong covalent bonds between the two layers formed under UV curing. The conductive tracks produced using this design have an excellent waterproofing property and good flexibility.

Future work will explore a flexible and stretchable silver paste to replace the silver paste used in this study in order to produce washable conductive tracks on textile with a target of surviving up to 50 domestic washes.

Footnotes

Acknowledgment

The authors would like to thank Klopman International for supply of the fabric used in this study.

Funding

This work was supported by the EU Framework 7 NMP Project MICROFLEX (Grant number CP-IP211335-2).

References

Bouwstra S, Wei C, Feijs L, et al. Smart jacket design for neonatal monitoring with wearable sensors. In: 6th international workshop wearable and implantable body sensor networks, 2009, pp.162–167.

Park

Jayaraman

. Enhancing the quality of life through wearable technology. IEEE Eng Med Biol Mag 2003; 22: 41–48.

Engin

Demirel

Engin

. Recent developments and trends in biomedical sensors. Measurement. 2005; 37: 173–188.

Guo L, Berglin L and Mattila H. Improvement of electro-mechanical properties of strain sensors made of elastic-conductive hybrid yarns. Textil Res J. Epub ahead of print 11 July 2012. DOI: 10.1177/0040517512452931.

Post

Orth

Russo

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

Gimpel

Möhring

Müller

. Textile-based electronic substrate technology. J Ind Textil 2004; 33: 179–189.

Lam Po Tang

. Recent developments in flexible wearable electronics for monitoring applications. Trans Inst Meas Control 2007; 29: 283–300.

Locher

Tröster

. Screen-printed textile transmission lines. Textil Res J 2007; 77: 837–842.

Karaguzel

Merritt

Kang

. Utility of nonwovens in the production of integrated electrical circuits via printing conductive inks. J Textil Inst 2007; 99: 37–45.

10.

Karaguzel

Merritt

Kang

. Flexible, durable printed electrical circuits. J Textil Inst 2009; 100: 1–9.

11.

Kim H, Kim Y, Kim B, et al. A wearable fabric computer by planar-fashionable circuit board technique. In: sixth international workshop on wearable and implantable body sensor networks (BSN). 2009, pp.282–285.

12.

Kim

Yoo

H-J

. Electrical characterization of screen-printed circuits on the fabric. IEEE Trans Adv Packag 2010; 33: 196–205.

13.

Kazani

Hertleer

De Mey

. Electrical conductive textiles obtained by screen printing. Fibres Textil East Eur 2012; 20: 57–63.

14.

Kazani I, Hertleer C, De Mey G, et al. Dry cleaning of electroconductive layers screen printed on flexible substrates. Textil Res J. Epub ahead of print 26 June 2012. DOI: 10.1177/0040517512449050.

15.

Bai

Zhang

Dai

. Water resistance of the membranes for UV curable waterborne polyurethane dispersions. Prog Org Coat 2007; 59: 331–336.

16.

Smart Fabric Inks Ltd. www.fabinks.com (accessed 15 December 2012).

17.

Torah R, Yang K, Beeby S, et al. Screen-printed multilayer meander heater on polyester cotton. In: Textile Institute world conference, Malaysia.

18.

Xin

Shen

. Novel cationic polyurethane-fluorinated acrylic hybrid latexes: synthesis, characterization and properties. Colloid Surf A 2011; 384: 205–211.

19.

Żenkiewicz

. Methods for the calculation of surface free energy of solids. J Achievements Mater Manuf Eng 2007; 24: 137–145.