Textile Research Journal OnlineFirst, published on October 13, 2009 as doi:10.

1177/0040517509348331

Textile Research Journal

Article

Advances in Topographic Characterization of Textile Materials

All textile materials, having periodic surfaces, show horizontal and vertical repetitive unities. For this reason, different length scales have to be taken into account by interpreting topographic data measured. In this study, a topographical characterization method for textile materials at different length scales is presented and justified. The topographical study of textile materials using different length scales permits us to characterize the surfaces considering their specific morphologies due to the type of weave, yarn and filament/ fibers separately. The use of a scale concept to characterize textile surfaces seems to be a new skill that helps to correlate textile parameters, topography, and topographical changes with interface phenomena such as spreading, wetting, capillary penetration, and soil release.

Abstract

Alfredo Calvimontes1, Victoria Dutschk and Manfred Stamm

Leibniz Institute of Polymer Research Dresden, Hohe Strasse 6, 01069 Dresden, Germany

Key words textiles, roughness, different length

scales

Recent studies [13] have shown that textile construction parameters, such as fineness of filaments and yarn, warp and weft density as well as the type of weave, control the texture, surface topography, and morphology of fabrics. Fabric topography affects the porosity and strongly influences the textile characteristics such as fabric mass, thickness, draping ability, stressstrain behavior, or air permeability. Moreover, there are significant differences between the soiling behavior and soil release of textile materials with different topographic structures despite the similarity of their chemical nature [46]. In the studies cited above, a large number of roughness and waviness parameters were obtained that did not take into account the scaled-morphologic periodicity of each surface studied and its influence on the whole topography. All textile materials having periodic surfaces show some horizontal and vertical repetitive patterns; therefore, different length scales have to be taken into account for a proper interpretation of the topographic data measured [7].

Materials
Polyester fabrics of three different types of weave (Figure 1) were manufactured at the Institut fr Textil- und Bekleidungstechnik (ITB) at the Technische Universitt Dresden using filaments produced by spinning of the same polymer material (polyethylene therephthalate). Warp yarn was formed from flat filaments, while weft yarn was textured by three different processes (Figure 2, Table 1).

Experimental methods
Topography measurements
An imaging measuring instrument was used for the optical analysis of the topography of textile materials, MicroGlider

1 Corresponding author: tel: +49 351 4658 212; fax:+ 49 351 4658 474; e-mail: calvimontes@ipfdd.de

Textile Research Journal Vol 0(0): 112 DOI: 10.1177/0040517509348331

The Author(s), 2009. Reprints and permissions: http://www.sagepub.co.uk/journalsPermissions.nav

TRJ

Textile Research Journal 0(0)

Figure 1 Polyester fabrics used in the present study.

Table 1 Textile parameters of filaments.

Yarn Warp Weft
a

Number of filaments 128 128

Structure flat textured, tangled

Filament fineness [dtex]a 0.78 0.92

Filament diameter [m] 6 7.5b

Yarn fineness [tex]a 19.9 11.5

Measured according to DIN 1973.1996; b before texturing.

(FRT, Germany). Unlike conventional microscopes, which simultaneously image all the points in the field of view and capture a 2D image, a chromatic confocal microscope records only one object point per given unit of time. The field measured is constructed by xy scanning. This novel

Figure 2 Microscopic images of warp yarns (a), weft yarns (b), warp filaments (c) and weft filaments (d).

optoelectronic setup based on a quasi confocal, z-axis extended field, was developed for high-resolution noncontact 3D surface metrology, including roughness characterization and surface flaw detection. The instrument uses a chromatic white-light sensor (CWL), based on the principle of chromatic aberration of light [8]. As can be seen in Figure 3, white light is focused on the surface by a measuring head with a strongly wavelength-dependent focal length (chromatic aberration). The spectrum of the light scattered on the surface generates a peak in the spectrometer. The wavelength of this peak along with a calibration table reveals the distance from the sensor to the sample. The sensor works on transparent, highly reflective or even matt black surfaces [9]; it is extremely fast and has virtually no edge effects. The instrument used allows a lateral measure range and a vertical measure range up to 100 mm and 380 m, respectively, and a lateral resolution and vertical resolution up to 1 m and 3 nm, respectively. In [10,11], the use of chromatic confocal microscopy to measure topography of textile surfaces was compared with the use of high-resolution scandisk confocal microscopy (SCDM). According to these studies, wider cut-off lengths and larger z-ranges make chromatic confocal imaging

Advances in Topographic Characterization of Textile Materials A. Calvimontes et al.

TRJ

The use of tables that relate predicted values of the mean rough height (Rz), the root mean square roughness (Rq), and arithmetic mean roughness (Ra) to Lm is frequently recommended to set the optimal value of Lm for periodic as well as non-periodic surfaces. As optimal sampling conditions are strongly dependent on the type of material to be characterized, researcher experience is usually required. For this reason, a systematic procedure to define optimal cut-off length and resolution values is proposed: 1. Acquiring topographical data at the highest resolution available (minimal value of x) using different Lm values. Here two different procedures are recommended: (a) only one measurement at the highest Lm and posterior zooming (sub-area extractions), or (b) independent measures using the same zero point position. 2. The use of statistical criteria in order to define an optimal value of Lm by analyzing Wz, Rz, and Ra curves as functions of Lm. 3. Acquiring topographical data with the defined optimal Lm using different values of resolution. 4. The use of statistical and topographical criteria to analyze Rz and Ra as functions of x, in order to define an optimal resolution. Woven plain and twill polyester fabrics were used to probe the presented procedure concerning optimization of sampling conditions. Figure 4 shows that the waviness of woven plain fabrics is statistically reliable above Lm = 2 mm, but in the case of twill fabric, the optimal cut-off length has to be higher than 3 mm. However, this reasoning takes only into account the statistical behavior of Wz. By applying fast Fourier transform (FFT) filtering [15], the waviness images calculated, as illustrated in Figure 5, show that almost constant values of Wz, as shown in Figure 4, correspond in both cases to a quantification of plane irregularities (wrinkles) of the fabrics. If the characterization is aimed at the morphology quantification caused by the fabric structure, the calculation of the waviness has to be realized using Lm in the range of 0.5 to 1 and 1 to 2 mm for woven plain and twill fabrics, respectively. In the characterization case aimed at the study of topography of the complete fabric surface without regarding the fabric structure (morphology), the recommended Lm value is 3 mm for both fabrics, since over this cut-off length the values of Rz and Ra remain approximately constant (cf. Figure 4). By Lm = 3 mm, different resolution values were used to obtain new topographical data. Figure 6 shows that a lateral resolution of 4 m is enough to produce reliable information concerning Rz. On the other hand, values of Ra and their correspondent standard deviation (cf. Table 2) show

Figure 3 Schematic presentation of the measuring principle of chromatic confocal microscopy.

more appropriate than SCDM to measure topographical characteristics of polyester and cotton fabrics. Depending on the fabric characteristics, and the structure and size of repetitive units, other non-contact measurement methods (scanning electron microscopy, Confocal laser scanning microscopy, Confocal Scanning Optical Microscopy (CSOM), Conoscopy holography (CSL), etc.) can be used. It is absolutely necessary that a combination of cut-off length, zrange, and resolution have to provide statistically representative topographical data. It is important to note that the selection of a method due to its high resolution could be inadequate if the cut-off length available or z-range is too small. On the other hand, the use of a very high resolution and larger cut-off lengths (scan areas) results in data whose excessive size could demand extremely long calculation times and special or non-existent hardware and software.

Calculating Optimal Sampling Conditions

Cut-off length (Lm), defined as the length of one side of the square sampling area, and resolution (distance between measured points x, assuming that x = y) are the most important sampling parameters, which apart from particular instrumental dependent parameters, such as light intensity, measuring frequency, etc., have to be optimally defined before characterizing topography. Tsukada and Sasajima [12] and Yim and Kim [13] discussed the problem of an optimum sampling interval (Lm) by checking the variance of the root mean square roughness (Rq) for a surface under different sampling intervals. According to Stout et al. [14], the recommendation mentioned above for the choice of sampling interval is doubtful because of the fact that optimum Lm seems to influence the amplitude parameters (wave height Wt and waviness Wz).

TRJ

Textile Research Journal 0(0) tions of Ra for woven plain and twill fabric obtained using Lm = 1 mm are 0.51 and 0.62, respectively, which are clearly much more dependent on resolution.

Topographic Characterization
As shown in Figure 5, topographic parameters obtained by FFT filtering provide different types of information depending on the cut-off length used. Due to the structural diversity of textile materials, their classification by unit size and morphology on different length scales is necessary. A suggestion to find the general range values of Lm in order to identify the different measurable length scales is not reasonable. However, the specification of at least three different length scales (macro-, meso- and micro-scale) is absolute necessary to describe morphologically the homogeneous textile groups. From a conceptual point of view, each one of the length scales proposed for a textile structure has to provide specific information about the surface morphology and topometry of the materials. In all cases, the highest available lateral resolution by confocal chromatic imaging (x =1 m) was used, while the value of Lm was the parameter used to determine the length scale to be studied.

Results and Discussion

Textile Macro-topography
Figure 4 Topographic parameters for polyester fabric surfaces as functions of Lm. Macro-morphological irregularities of textile surfaces such as folds and wrinkles can be studied using FFT filtering of topographical data measured by large values of Lm. A cutoff length value larger than 3 mm was suggested above to quantify the plane irregularities (waves and wrinkles) of the fabrics. In order to provide the macro-topographical information desired, according to the large amount of topographical data measured for three different types of woven polyester textile surfaces (plain, twill, and Panama), a total measured area (Lm Lm) should cover at least 169 (132) repetitive units (r.u.). In 3D-waviness diagrams Wz values (Figure 7) can be used as macro-morphological parameters, especially to characterize changes that occurred after treatment. In this case, the characterization has to be accompanied by quantification of the 2D relaxation/shrinkage (cf. ISO 5077:2007) that occurred while wetting or during mechanical processes, or by a combination of both. Dimensional changes (relaxation/shrinkage) of fabrics on a macro-scale influence their meso- and micro-topography due to the modification of the repetitive unit dimensions and, therefore, the distances between yarns, filaments, and fibers [16]. In Figure 7, only the macro-waviness diagram for woven plain fabric does not show any morphological influence of repetitive unit morphology; in this case, r.u. > 132. For

Table 2 The Ra values of the studied fabric by Lm = 3 mm are nearly independent of the resolution up to x = 30 m.

x [m]
woven plain 1 2 3 4 5 30 15.67 15.69 15.79 15.68 15.79 15.70

Ra [m] twill 17.24 17.26 17.39 17.23 17.38 17.17

= 0.056

= 0.070

that, using Lm = 3 mm, the arithmetic mean roughness hardly depends on resolution. However, standard devia-

Advances in Topographic Characterization of Textile Materials A. Calvimontes et al.

TRJ

Figure 5 Waviness images of polyester fabric surfaces as a function of Lm.

macro-topographical characterization of twill and panama types, their optimal Lm values have to be larger than 5 mm.

Textile Meso-topography
The meso-scale of textile materials should be used to describe the surface topography produced by the type of weave and yarn used, without considering the previously defined macro-topographic irregularities and details corresponding to fibers or filaments. A study of fabric surface topography on a meso-scopic scale using FFT filtering starts with the selection of a new optimal Lm value, which basically depends on the size of the fabric repetitive unit. From the large amount of experimental data for polyester fabrics studied, it was revealed that a sample area (Lm Lm) has to cover about eight repetitive units (Table 3). Another way to construct meso-topographic diagrams is the use of digital surface filtering, which calculates the arithmetical mean of each data point within its neighborhood [14,15,17]. The filter density used depends on the

Figure 6 Mean roughness of polyester fabric surfaces as a function of x.

TRJ

Textile Research Journal 0(0)

Figure 7 2D images (left) and 3D waviness diagrams (right) with Wz values for woven plain (a), twill (b) and panama (c) polyester fabrics.

fabric characteristics and has to be able to produce a surface without topographical details of fibers or filaments. Figure 8 shows the construction of meso-topographic surfaces by using FFT filtering and smooth filtering. In order to compare the morphologies and Wz values obtained, Lm and the filtering method used should remain the same during the characterization process. An application of the study on these length scales claims to know the relative z-distances between warps and wefts for woven fabrics and the amplitude of their wave (sinoidal) trace. As shown in Figure 8, wefts describe almost a linear trace (their amplitudes are small). As a consequence, the first contact of any solid with the fabric surface takes place by the warps (hills) and the final pen-

Table 3 Lm that contain about eight repetitive units are optimal to characterize the meso-topography.
Unit area [mm2] Woven plain Twill Panama 0.126 0.396 0.478 Lm [mm] 1.0 1.8 2.0 r.u. 7.94 8.18 8.37

etration of fluids on the fabric surface takes place principally on wefts (valleys). This finding plays a crucial role in understanding the wetting behavior of textile materials [16]. The application of volumetrical characterization criteria allows us to investigate the topographical conditions for fluids spreading over the surfaces (a connection between meso-morphology and spreading was presented in [16]). A height value (hv) can be found to divide all data points into two groups: those forming mountains and those forming canals between the mountains. There are two conditions for hv: (i) to be as small as possible and (ii) that the canals be connected in order to allow flow in all possible directions. In the case of twill and panama fabrics, due to their more anisotropic morphology, independent but endless canals can be formed. In the case of woven plain structures, all canals are connected to each other (Table 4).

Textile Micro-topography
Unlike macro- and meso-scales, characterization at a micro length scale reveals the influence of filament and fiber characteristics on the resulting topography. Profile, fineness, as well as the natural or machined texture of these elements or

Advances in Topographic Characterization of Textile Materials A. Calvimontes et al.

TRJ

Figure 8 Meso-topography of different polyester fabrics by FFT filtering and smooth filtering.

Table 4 Volumetrical characterization parameters by dividing the z-range according to the hv criteria. Smooth filtering (80%) was used to obtain the meso-topography.
plain twill Panama

Lm [mm] Wz [m] hv [m] Canals z-range (hv/Wz) [%] Canals area [%] Canals volume [m3/m2]

1 64.41 31.94 49.6 58.0 6.787

1.8 78.73 46.66 59.3 66.4 10.083

2 96.76 41.36 42.7 40.5 5.300

TRJ

Textile Research Journal 0(0)

Figure 9 Optimal Lm values for the characterization of warps and wefts micro-topography separately at the surface of woven plain polyester fabric.

Figure 10 Micro-topographical images of a warp and a weft. The elimination of micro-waviness was possible by FFT filtering.

the distances between them are only some of the possible characteristics which define the resulting morphology and topometry at this length scale. The selection of an optimal cut-off length in this case no longer depends on some statistical or mathematical criteria as seen at macro and meso length scales, but rather on the size and location of the set of filaments or fibers by type and orientation. To study the micro-topography of woven plain fabrics, warps and wefts should be zoomed separately (sub-area extraction). Optimal Lm values of warps and wefts depend on the type of weave and construction parameters such as yarn types, their diameters, warp densities, weft densities, etc. Depending on textile structure, more than one Lm value could be necessary for a complete micro-topographical characterization, as shown in Figure 9.

The number of sub-areas to be isolated depends on the topographical parameters studied and on the standard deviations of their mean values. Usually five different zooms should be enough to characterize polyester monofilament fabrics. Depending on the characterization criteria, the elimination of micro-waviness, a consequence of yarn profile and fabric meso-topography, is possible by FFT filtering, as shown in Figure 10. Using the new topographical data generated, it is possible to calculate any micro-topographical parameter by profiling or by using the whole surface. The volumetrical characterization is a good tool to measure textile surfaces through the evaluation of porosity or filling quantities at different deep heights. The calculation of the skewness, kurtosis or surface relative smooth (SRS) [10] can be important

Advances in Topographic Characterization of Textile Materials A. Calvimontes et al.

TRJ

Figure 11 Three different wetting regimes for a textile surface.

Figure 12 Dynamic wetting: meso-morphology controls the spreading rate of a liquid drop on a textile structure.

for studies of surface modification treatments such as by heat setting or plasma modifications. Fractal dimension or Wenzel roughness factor could be of interest to characterize micro-topographical modifications of natural fibers, e.g. changes caused by enzymatic action.

Example of Application 1: Meso-topography and Spreading

In [10,16], 14 different polyester fabrics having plain, twill, and panama structures were characterized to show how the use of topographic characterization at different scales can provide important information of the spreading behavior. On the basis of macroscopic water drop base changes measured with a dynamic contact angle tester (Fibro DAT 1122, Fibro System, Sweden), the wetting behavior of a water drop can be divided into three regimes (Figure 11): dynamic wetting, defined as growing of the drop diameter depending on time (also known as spreading); quasi-static wetting, where the drop diameter remains approximately constant; and penetration, which is marked by liquid drop absorption into fabrics depending on time. By using the waviness as meso-topographical parameter, it is evident that the meso-topography of the fabrics controls the spreading rate of a liquid drop (Figure 12). For the plain weave, an increase of the waviness depth causes a decrease of the spreading rate; warp yarns (hills) slow down the liquid motion (Figure 13). For twill weave, an increase of the waviness depth causes formation of deep and long domains of weft yarns (canals) with small islands. As a consequence, an increase of the spreading rate is observed. Finally, for panama weave, an increase of the waviness depth causes for-

mation of long and quasi-endless (without islands) deep domains (canals). Consequently, the waviness depth and spreading rate are proportional to each other. A thorough comparison between topographic parameters for the 14 fabrics having three different types of weaves reveals that the respective morphology at a meso length scale controls the spreading rate.

Example of Application 2: Micro-topography, Wetting and Cleanability

At a smaller scale, by zooming of warps and wefts separately, topography measurements and characterization at different length scales provide important information about changes in textile microstructures. Using this information, the behavior of a liquid drop on a fabric surface while wetting can be explained. In [10,16], on the basis of experimental results, revealing differences in three basic types of woven fabrics plain, twill, and Panama in respect to capillarity and water penetration (Figure 14), the concept of a novel wicking model was developed. This conceptual model was verified in respect to the cleanability behavior of paraffin oil and acetylene black soils. The results illustrated in Figure 14 show that: (i) warp yarn topography hardly affects the cleanability; (ii) spaces between fibers make the plain weave surface oleophil (the larger they are, the more stains penetrate); (iii) spaces between fibers make the twill weave surface oleophob. The larger and deeper they are, the more stain penetrates and the worse their cleanability and (iv) the weft yarn roughness controls the hydrophobicity or hydrophilicity of fabrics and, as a consequence, their cleanability (cf. Figure 15).

TRJ

10

Textile Research Journal 0(0)

Figure 13 Respective textile morphology at a meso-length scale controls the spreading rate. Above: morphology; below: spreading directions of a liquid drop.

Figure 14 Liquid flow in the warp and weft directions occurs by two different regimes, depending on micro-topography.

Conclusion
The topographical study of textile materials using a length scale concept allows us to characterize surfaces separately by considering and analyzing their specific morphologies caused by the type of weave, yarn, and filament/fibers. With the resulting information, the correlation of topography and topographical changes due to modification processes, such as heat setting, impregnation, and wash dry cycles with interface phenomena such as spreading, wetting, capillary actions, and cleanability, is more specific and explanatory.

Acknowledgements
This research is financially supported by Sasol Germany GmbH. The authors are grateful to Dr Beata Lehmann and Dr Birgit Mrozik (both from the Institut fr Textil- und Bekleidungstechnik, Dresden) for providing textile materials.

Advances in Topographic Characterization of Textile Materials A. Calvimontes et al.

11

TRJ

Figure 15 The conceptual model proposed in [16] allows a better understanding of the cleanability phenomenon of polyester fabrics by using a different length scale concept for their characterization. Soil material: paraffin oil and acetylene black in the ratio 97.98:2.02.

References
1. Calvimontes, A., Synytska, A., Dutschk, V., Bell, and Ch., Lehmann, B., Einfluss der Struktur von Polyestergewebe auf ihre Benetzbarkeit, Melliand Textilberichte, 1(2), 6467; Melliand English, 1(2), E16E18 (2006). Dutschk, V., Myat, S., Mrtin, J., Stolz, M., Breitzke, B., Cherif, Ch., and Heinrich, G., A Comparative Analysis between Different Ether Carboxylic Acids with Respect to Wettability and Surface Topography of Abrasively Treated Polyester Fabrics, Tenside Surf. Det., 44(6), 248254 (2007). Hasan, M. M. B., Calvimontes, A., Synytska, A., and Dutschk, V., Effects of Topographic Structure on Wettability of Differently Woven Fabrics, Textile Res. J., 78(11), 9961003 (2008). Calvimontes, A., Dutschk, V., Breitzke, B., Offermann, P., and Voit, B., Soiling Degree and Cleanability of Differently Treated Polyester Textile Materials, Tenside Surf. Det., 42(1), 1722 (2005). Calvimontes A., Dutschk, V., Koch, H., and Voit, B., New Detergengy Aspects through Visualisation of Soil Release Polymer Films on Textile Surfaces, Tenside Surf. Det., 42(4), 210216 (2005). Hasan, M. M. B., Dutschk, V., Calvimontes, A., Horrmann, G., Heinrich, G., and Cherif, Ch., Influence of the Cross-sectional Geometry on Wettability and Cleanability of Polyester Woven Fabrics, Tenside Surf. Det., 45(5), 27479 (2008). Hasan, M. M. B., Calvimontes, A., and Dutschk, V., Correlation between Wettability and Cleanability of Polyester Fabrics Modified by Soil Release Polymer and their Topographic 8.

2.

9.

3.

10.

4.

11.

5.

12. 13.

6.

7.

14.

Structure, J. Surfacts Detergts, DOI 10.1007/211743-009-1130-x (2009) http://www.springerlink.com/content/a88188k024169110/ fulltext.html. Ruprecht, A. K., Krner, K., Wiesendanger, T. F., Tiziani, H. J., Osten, W., and Lcke, P., Chromatic confocal sensors for micro-topography measurements, 50, Internationales Wissenschaftliches Kolloquium der TU Ilmenau: Maschninenbau von Mikro bis Nano, Technische Universitt Ilmenau, September 2005. Ruprecht, A. K., Krner, K., Wiesendanger, T. F., Tiziani, H. J., and Osten, W., Chromatic Confocal Detection for High Speed Micro-topography Measurements, Proc. SPIE, Volume 3302, pp. 5360 (2004). Calvimontes, A., Topographic Characterization of Polymer Materials at Different Length Scales and the Mechanistic Understanding of Wetting Phenomena, PhD Thesis, Technische Universitt Dresden, Germany, 2009. Lukesch, M., Vergleichende Untersuchungen mit modernen optischen Messmethoden zur Bestimmung der Mikrotopographie von Oberflchen, Praxissemesterarbeit, Hochshuleu Zittau/Grlitz (FH) University of Applied Sciences, Germany, 2009. Tsukada, T., and Sasajima, K., An Optimum Sampling Interval for Digitizing Surface Asperity Profiles, Wear 83, 119128 (1982). Yim, D. Y., and Kim, S. W., Optimum Sampling Interval for Ra Roughness Measurement, Proc. Instn. Mech. Engrs., 205: 139142 (1991). Stout, K. J., Sullivan, P. J., Dong, W. P., Mainsah, E., Luo, N., Mathia, T., and Zahouani, H., The Development of Methods for the Characterisation of Roughness in Three Dimensions,

TRJ

12

Textile Research Journal 0(0)

und Modifizierung von Kunststoffoberflchen: NDVaK, 15. Neues Dresdner Vakuumtechnisches Kolloquium, Dresdner Transferstelle fr Vakuumtechnik e.V., October 2007, pp. 48 52. 17. Fries Research & Technology GmbH, Basis FoRT-Bildung 01, FRT Metrologie Seminar, Bergisch-Gladbach, October 2004.

Commission of the European Communities, Brussels-Luxembourg, 1993. 15. Raja, J., and Radhakrishnan, V., Filtering of Surface Profiles using Fast Fourier Transform, Int. J. Mach. Tool. Des. Res., 19, 133141 (1979). 16. Calvimontes, A., Dutschk, V., Cherif, C., and Heinrich, G., Ein neues Konzept zum besseren Verstndnis der Penetration von Flssigkeiten in textilien Oberflchenstrukturen, Beshichtung