**3. Results and Discussion**

#### *3.1. Electrical Characterization*

For the purpose of determining the electrical conductivity of the printings an approach based on measuring the electrical resistance was considered. A low resistivity indicates a material that readily allows electric current. Table 5 shows a summary of the printing results for all the fabrics in terms of electrical resistivity. Four different samples of each fabric were measured. The orientation of the measurements on the woven fabrics were separately considered. In addition, results are graphically shown in Figure 2.

**Table 5.** Electrical resistance of the printings. Measurements have been made both in warp and weft directions.


Regarding the influence of the textile structure on the electrical behavior shown in Figure 2a, both measures on warp and weft directions agree that the woven fabric that presents the higher electrical resistance, and therefore worse electrical conductivity, is the substrate coded as T8. T8 was characterized by a lower interlacing coefficient and the highest mass per unit area.

These results coincide with the conclusions about the surface properties of woven fabric on electrical performance through a screen-printed technique. The fabric substrate with the smallest pore size and roughness shows a higher printing precision and lower electrical resistance of screen-printed conductive lines [20]. Therefore, the surface structure of the fabric substrate determines to some degree, not only the printing precision of conductive lines, but its electrical properties as well.

**Figure 2.** Graphic representation of the electrical resistance of the printings ( ) measured on warp direction and ( )measured on weft direction. (**a**) Electrical resistance affected by the structural variations of the textiles; and (**b**) Electrical resistance affected by the material composition of the textiles.

In order to address these challenges, surface pre-treatment onto rough and porous substrates or coatings and lamination processes should be performed in order to produce a continuous conductive pathway onto the textiles [5,10].

In order to deeply explore the relation found between woven fabric density and conductivity, results were plotted, as shown in Figure 3.

**Figure 3.** Graphic representation of the relation between electrical resistance of the printings measured on warp direction ( ) and the mass per unit area ( ).

According to results shown in Figure 3, with regard to the influence of the textile structure on the electrical behavior, it could be expected that the higher the mass per unit, the lower conductivity due to the ink expansion through the inner part of the textile.

Moreover, Figure 2b allows the analysis in terms of the impact of the textile composition on the electrical behavior of electronic printing through the flexographic technique. Again, resistance measures in both directions agree and show that the woven fabric T10 presents higher electrical resistance. Even though the measure on the warp direction shows

better electrical conductivity, it is worse than the other textile samples made from other materials. The weft of T10 was made from cotton. This makes sense, because of the moisture absorption capacity of cellulosic fiber. In this way it can be demonstrated that cotton performs worse as the weft material for printing electronics with the flexo technique.

An alternative solution for improving conductivity on printed electronic textiles with similar problems would be pre-treatment of the surface substrates by applying different chemical compounds that increase the adhesion of the ink avoiding its absorption.

#### *3.2. Physical Characterization*

Focusing on analyzing the print uniformity of each layer of the electronic flexo printed woven textiles, optical macroscopic images were taken with 20 magnifications. In parallel, in order to discuss the ink penetration and the adhesion in each substrate, high-resolution topographic images were taken with Scanning Electron Microscope Field Emission Gun (FE-SEM).

Table 6 shows the images obtained for the Optical and FE-SEM characterization of the woven substrates with different textile structures. The FE-SEM (×100) images show a visual characterization of the fabric and ink, and the FE-SEM (×150) images on the right show maps of crystalline and textural orientations by EBSD for a determination of the position of the silver particles.


**Table 6.** Optical and FE-SEM characterization (I).


**Table 6.** *Cont.*

Regarding optical results, images in Table 6 show a total color uniformity on the substrates' surface not depending on the textile structure. The silver shade obtained is

typical from conductive inks which contains silver nanoparticles. In addition, the images reveal the porosity of the woven fabrics as they show their characteristic textiles holes, being different for each structure. It should be highlighted the picture of the substrate T8 as it is the one with more holes between the yarns forming the woven fabric, in comparison with the other samples.

In respect of FE-SEM results, thanks to images of the cross section of the substrates shown in Table 6, it can be observed how weft threads with different titles and densities are intertwined with the warp threads. These pictures reveal the enclaves formed by the interlacing of the threads in both directions, obtaining a roughness of the fabric and presenting different heights. Depending on the structure and yarn title, the enclave formed is different and it directly affects to the fabric thickness. As already stated on the fabrics' characterization in Table 2, the FE-SEM (×100) images in Table 6 confirm the thickness is greater for fabrics whose ligament is sarge (substrates T5–T8).

Furthermore FE-SEM (×150) results with more magnification allow to observe the ink penetration and distribution in the inner part of the textile. Images demonstrate that there are particles of the silver ink between the fibers of the weft threads.

Meanwhile, Table 7 shows the images obtained for the Optical and FE-SEM characterization of the woven substrates with different materials on the weft.

**Table 7.** Optical and FE-SEM characterization (II).

Regarding ink uniformity, all the woven fabrics present a good visual level at the print in spite of the porous surface. To ensure this uniformity in printing electronics with low viscosity inks onto a rough and porous textile surface is a great challenge, due to the orientation of fibers or yarns and the change of fiber morphology constantly [5].

With respect to ink penetration, the images in Table 7 show that ink does not remain at the surface, but penetrates until the lower side of the textile. Thus, results coincide with the relation pointed out in Figure 3 about the influence of the mass per unit area on the electrical behavior. Whereas conventional printing that uses thickeners to increase the viscosity of the paste achieve deposition only on the surface, the low ink viscosity used in flexo printing penetrates to the inner of the textile, the interior of the textile. Thus, the topography of the sample plays an important role in achieving the homogeneity and continuity of the ink. Therefore, it is demonstrated that the higher the mass per unit, the lower conductivity due to the ink expansion through the inner part of the textile.

In terms of the impact of the material of the weft, the images shown in Table 7 do not show significant differences among them.

The color loss and staining resulting from desorption and/or abrasive action of the samples was evaluated according to EN ISO 105-C06:2010 tests for color fastness [29]. The grade of color fastness to domestic and commercial laundering is presented in Tables 8 and 9.


**Table 8.** Color fastness results (I).



For assessing the change in color, woven treated textiles were compared with a grey scale complying with ISO 105-A02. Regarding assessing staining, the grey scale used follows ISO 105-A03. Both scales consist of five pair of non-glossy grey color chips which illustrate the perceived color differences corresponding to fastness rating 5, 4, 3, 2 and 1. A range of 5 is given only when there is no perceived difference between the tested specimen and the original material.

According to results shown in Table 8, obtained values for change in color and staining are between 4 (Very Good) and 5 (Excellent). Therefore, it can be concluded that structural variations have not a relevant influence on the color fastness of the flexo electronic printings.

However, results shown in Table 9 demonstrate that color fastness decreases with natural fibers such as cotton and wool. The reason can be that ink adhesion improves in in synthetic fibers such as polyester thanks to the curing time after the flexography printing using 150 ◦C, as this type of fiber is a thermoplastic material and being its glass transition temperature around 70 ◦C [30].

In order to address these challenges in woven textiles with natural fibers, surface pretreatment should be done in order to improve the ink adhesion and therefore the electrical behavior [5]. It has been proved that special pretreatment processes on the fabric substrates

improve the wash fastness for other printing techniques such as digital printing [31] or ink-jet [32]. According to a previous review [33], an increment of ink volume improves the ink coverage, upgrading in this case the conductivity, nevertheless it enhances the ink wash-out effect. For this reason, the ink volume transferred to the substrate should be optimized when conductivity and color fastness to washing are the objectives. In addition, coating and lamination processes could be done to ensure the continuous conductive pathway on textiles [10].

Spectrophotometric methods are adequate and objective for determining the color values of the fabric surface [34]. To assess variations found on the printings, measurements have been performed according to ISO 105-J01:1997 General principles for measurement of surface color [35].

The CIELAB, or CIE L\* a\* b\*, color system represents quantitative relationship of colors on three axes: L\*value indicates lightness, and a\* and b\* are chromaticity coordinates. It is the most widely used method for measuring and ordering object color. The results of the analysis of color of the printings onto the woven fabrics are shown in Table 10. For a better understanding, results have been plotted and can be consulted at the Figure 4. On the color space diagram, L\* is represented on a vertical axis with values from 0 (black) to 100 (white). The a\* value indicates red-green component of a color, where +a\* (positive) and −a\* (negative) indicate red and green values, respectively. The yellow and blue components are represented on the b\* axis as +b\* (positive) and −b\* (negative) values, respectively. At the centre of the plane is neutral or achromatic [36].

**Figure 4.** Two-dimensional CIELAB (+L\*, +a \*, +b\*) coordinates representation. (**a**) CIELAB lightness (+L\*) of textile structure variations T1–T8; (**b**) CIELAB a\* and b\* coordinates of textile structure variations T1–T8; (**c**) CIELAB lightness (+L\*) of textile material variations T9–T11; (**d**) CIELAB a\* and b\* coordinates of textile material variations T9–T11.


**Table 10.** CIELAB\* coordinates (D65/10◦).

Lower L\* indicates that the sample becomes darker, and all printed samples show low light (L\*) values near to 50. It is observed that there is no correlation between the conductivity and the tone obtained from the coordinates a\* and b\* in contrast to silver coatings of polymeric substrates, where it is demonstrated that s shifting in the a\* and b\* during sintering means best conductivity [37]. On the other hand, it is observed that the sample T8/T9, the one with more weight and less conductivity, shows the lowest L\* value, being darker than the other samples. This fact corroborates firstly, that variations on the fabric density have an influence on the color values [34], and secondly that the penetration of the silver particles into the textile, when the silver particles remain on the surface, thanks to emulsion resins, present values of L\* between 60–70 [38].

#### **4. Conclusions**

It was proved that a direct relationship exists between the electrical performance of a flexographic printed textile with the structure of the woven as well as with the material composition of the weft.

After performing controlled flexographic printing processes onto woven fabrics by using a silver electrical ink, physical and electrical analysis led to the following conclusions. On the one hand, from a perspective focused on the structural parameters, the weave construction, interlacing coefficient, yarn count and fabric density, there was an effect due to the higher mass per unit of the fabric and lower conductivity due to the ink expansion through the inner part of the textile. On the other hand, based on the weft material composition (cotton, polyester, and wool), it was shown that cotton performs worse as weft material for printing electronics with the flexo technique due to the moisture absorption capacity of cellulosic fiber.

It is believed that these findings will provide some important support for printing electronic devices on woven textiles by using the flexographic technique.

The study's next steps will consist of improving the conductivity of flexographic printed textile by carrying out surface pre-treatment, coating and lamination processes that increase the ink adhesion and therefore the electrical behavior.

#### **5. Patents**

FERRI, J.; RODES-CARBONELL, A.M. and MORENO, J. (2020). Dispositivo NFC flexible para la medición, almacenamiento y transmisión de datos (Spain U202130440) http:// www.oepm.es/pdf/ES/0000/000/01/26/48/ES-1264864\_U.pdf (accessed on 12 April 2021).

**Author Contributions:** Conceptualization, J.F., E.G.-B., I.M., E.B.-B. and A.M.R.-C.; methodology, E.G.-B., I.M., E.B.-B. and A.M.R.-C.; validation, J.F., E.G.-B., I.M., E.B.-B. and A.M.R.-C.; formal analysis, A.M.R.-C.; investigation, J.F., E.G.-B., I.M., E.B.-B. and A.M.R.-C.; writing—original draft preparation, E.G.-B., I.M., E.B.-B. and A.M.R.-C.; writing—review and editing, E.G.-B., I.M., E.B.- B. and A.M.R.-C.; supervision, E.G.-B., I.M. and E.B.-B. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work was supported by the Spanish Government/FEDER funds [Ministerio de Economía y Empresa (MINECO)/Fondo Europeo de Desarrollo Regional (FEDER)] under Grant RTI2018-100910-B-C43.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The data presented in this study are available on request from the corresponding author.

**Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

#### **References**

