Determining Tactile Comfort of Cellulose-Based Woven Fabrics, Knitted Fabrics and Terry Towels Using a Novel Instrument

ABSTRACT

In this study, tactile comfort of woven bed sheets, terry towels, and weft knitted fabrics made of cotton and blends was investigated. Tactile Sensation Analyzer (TSA) was used for determining surface characteristics and low-stress mechanical properties of fabrics. Micro and macro-surface variations were determined via a special sound analysis technique, and low-stress mechanical properties (deformation, elasticity, hysteresis, and plasticity) were measured in out-of-plane deformation state. Total hand (TH) scores of fabrics were determined by expert assessors via sensory tests. Findings of the study revealed that surface and low-stress mechanical properties measured using TSA were strongly and significantly correlated with sensory evaluation results in general. It was detected that tactile comfort of knitted fabrics and towels was strongly related to the magnitude of macro-surface variations; meanwhile, micro-surface variations were found out to be a more determinant parameter for bed sheets. It was observed that deformation and recovery characteristics of bed sheets and towels have a significant effect on TH scores as well. Hand of fabrics was also investigated by a conventional nozzle test equipment for comparison purposes, and it was detected that TSA results can represent fabric hand better than the raw data obtained from the nozzle test equipment.

摘要

在本研究中，对棉和混纺织物制成的机织床单、毛圈毛巾和纬编织物的触觉舒适性进行了研究. 采用触觉分析仪（TSA）测定织物的表面特性和低应力力学性能. 通过特殊的声音分析技术确定微观和宏观表面变化，并测量平面外变形状态下的低应力力学性能（变形、弹性、滞后和塑性）. 织物的手部（TH）总分由专家评估员通过感官测试确定. 研究结果表明，使用TSA测量的表面和低应力机械性能与总体感官评估结果密切相关. 研究发现，针织物和毛巾的触觉舒适性与宏观表面变化的大小密切相关; 同时，发现微表面变化是床单的一个更具决定性的参数. 观察到床单和毛巾的变形和恢复特性对TH评分也有显著影响. 为了进行比较，还通过传统的喷嘴测试设备对织物的手感进行了研究，发现TSA结果比从喷嘴测试设备获得的原始数据更能代表织物的触感.

KEYWORDS:

• Tactile comfort
• fabric hand
• sensory evaluation
• surface structure
• low-stress mechanical properties
• Tactile Sensation Analyzer

关键词:

• 触感舒适
• 织物手
• 感官评价
• 表面结构
• 低应力机械性能
• 触觉感知分析仪

Introduction

All textile products are made up of individual attributes such as aesthetics, quality, usefulness, and performance which are associated with the product’s features and benefits. Consumers may focus on one or more determinant attributes – or most likely several important attributes – during a particular purchase decision. Fabric hand has always been an important, and in certain cases, even a determinant attribute for textile industry. In this respect, being able to understand and measure tactile properties of textiles can provide significant advantages for producers.

In parallel with the demand for the development of devices to measure haptic properties of textiles, the studies in this field have gained momentum in recent years. Several researchers worked to understand the human tactile sensation and the cognitive mechanism of the brain on different components of fabric hand (Ciesielska-Wrobel and Langenhove 2012). Lee et al. (2007) worked with a glove-type measurement system to determine the characteristics of finger motion while evaluating the hand of a cloth. Tanaka et al. (2015) proposed a wearable tactile sensor for measuring skin vibrations. Chen et al. (2015) investigated the relationship between tactile perception and surface properties using an artificial finger. Zhang et al. (2016) recorded the brain potentials of panel members using the electroencephalogram (EEG) method during sensory evaluations. Wang et al. (2019a) conducted functional magnetic resonance imaging (fMRI) experiments to establish the relationship between tactile sensation and cerebral cortices of various brain areas.

It was previously proposed that three dimensions are enough to expound tactile sensation for most objects; hardness, roughness, and an uncertain dimension (Bolanowski et al. 1988; Chen et al. 2015). The hardness dimension represents fabric properties related to bending rigidity, whereas the roughness dimension refers to the surface characteristics. Researchers developed measurement systems such as Kawabata Evaluation System (KES-F), Fabric Assurance by Simple Testing (FAST), Comprehensive Handle Evaluation System for Yarns and Fabrics (CHES-FY), and Fabric Touch Tester (FTT) to evaluate the hand of fabrics by a series of objective measurements (Fan and Ng 2001; Kawabata and Niwa 1998; Liao et al. 2014; Zheng et al. 2021).

Several researchers proposed single measurement devices which evaluate fabric hand with a single testing process. Results given by these devices are complex combinations of hand-related parameters. Wool HandleMeter, PhabrOmeter, KTU-Griff-Tester, and several other test equipment are single measurement devices working according to a similar principle referred as ring method, nozzle test method, or extraction method, where a circular fabric sample is pushed or pulled through a ring or a nozzle, and the load-displacement graphs are used to predict fabric hand characteristics (Carrera-Gallissa, Capdevila, and Valldeperas 2014; El Mogahzy, Kilinc, and Hassan 2005; Kim and Slaten 1999; Pan, Lin, and Xu 2019; Pan and Yen 1992; Strazdiene and Gutauskas 2005; Strazdiene et al. 2009; Sun, Du, and Naebe 2018; Uren and Okur 2019; Wang, Mahar, and Hall 2012). Quick-intelligent handle evaluation system (QIHES-F) and equipment with parallel pins are other alternative single measurement methods (Iftikhar et al. 2021; Sun et al. 2019; Zhang et al. 2006). These devices are usually preferred by researchers for an overall evaluation of fabric hand. Touch Sensation Tester for Fabrics (TST-F) and Fabric Comfort Tester (FCT) are recently introduced devices which determine several sensorial properties of woven fabrics including mechanical, thermal, and acoustic comfort by realizing simultaneous tests (Hu, Chen, and Sun 2023; Liu, Wei, and Sun 2021).

Hand-related properties of fabrics can be determined independently via partial analysis as well. Low-stress mechanical properties such as extensibility and bending rigidity can be measured using standard methods and equipment (Hu, Li, and Yeung 2006). However, a small number of test instruments have been proposed for precise determination of in-plane shear behavior of fabrics (Bilisik 2012; El Abed et al. 2011; Uren, Oner, and Okur 2017). Unlike mechanical properties, investigating surface structure of fabrics can be quite complicated. Establishing a test method which would determine the surface characteristics of textiles has been a major interest of research (Liao et al. 2011). Measuring friction coefficient of fabric surface is a common method which is usually realized by sled test (Arshi, Jeddi, and Katbab 2012; El Mogahzy, Kilinc, and Hassan 2005; Jeddi et al. 2006; Kuo, Lin, and Su 2011; Van Amber et al. 2015). Measuring friction in skin-textile interactions and investigating surface characteristics of fabrics using a skin model or a tensioned wire were also proposed as alternative test methods (Bertaux, Lewandowski, and Derler 2007; Naebe et al. 2013; Pac et al. 2001; Ramsay, Fox, and Naylor 2012; Temel, Lloyd, and Johnson 2021).

Textiles may have very different characteristics depending on their production technology and end use. Moreover, prominent factors determining the hand of a textile product in a particular category could be different than another (Mahar, Wang, and Postle 2013). Early comfort studies focused on standardization of fabric hand and development of test methods for evaluating tactile comfort of woven fabrics (Kawabata and Niwa 1989, 1998). Despite the numerous research proposing methods to improve comfort of woven fabrics, most of the conventional woven fabrics still cannot reach the level of extensibility and softness provided by knitted fabrics, which are considered as the primary components of comfort perception (Behera, Ishtiaque, and Chand 1997; Luo et al. 2011; Nayak et al. 2009; Nazir et al. 2022; Sundaresan et al. 2016; Umair et al. 2023; Uren and Okur 2019). Because of their mentioned advantages, knitted fabrics are widely preferred in production of next-to-skin garments such as sportswear, undergarments, loungewear, leggings, and other casual clothing (Abu-Rous et al. 2018; Mahar, Wang, and Postle 2013; Wang, Mahar, and Hall 2012). Accordingly, the number of studies investigating comfort of knitted fabrics is on the increase (Atalie et al. 2021; Basra, Kumpikaite, and Asfand 2023; Jamshaid, Khan, and Ahmad 2023; Salman et al. 2022). Similar to next-to-skin knitted fabrics, towels are also in a regular contact with human skin, and tactile comfort is an essential element of towel quality as well (Kandzhikova and Germanova-Krasteva 2016; Kibayashi, Sukigara, and Yokura 2020; Singh, Behera, and Matsudaira 2014; Udurgucu, Sancar Besen, and Serafettinoglu 2024). Despite the growing interest, examining the comfort properties of knitted fabrics and towels remains challenging because of the special loop structure of knitted fabrics and pile structure of towels which create distinct differences in surface characteristics and mechanical properties. Since a great number of test instruments were mainly designed for measuring conventional woven structures, knitted fabrics and towels may fall outside the measurable range of many devices. Consequently, it may not be possible to precisely determine hand-related properties of these textile products using common test instruments and methods.

In parallel with the need to investigate comfort of fabrics with various structural properties, the demand for development of instruments which could measure tactile sensory properties of fabrics produced with different technologies is on the increase. Tissue Softness Analyzer is an instrument developed to measure surface variations and stiffness of tissue paper (Abu-Rous et al. 2018; Kim et al. 2021; Wang et al. 2019b, 2020). Even though originally designed to measure tissue paper, several researchers investigated the use of Tissue Softness Analyzer for evaluating surface and mechanical characteristics of woven and knitted textile structures. Abu-Rous et al. (2018) stated that Tissue Softness Analyzer can offer useful data for tactile comfort assessments of single jersey fabrics, and the results were reported to be comparable to other known test methods including Fabric Touch Tester, PhabrOmeter, and ring test equipment. Kim et al. (2021) investigated a total of 1001 types of fabrics including woven and knitted fabrics with different end uses (active wear, outer wear, dress and top). Macro and micro-surface variations of indicated fabrics were measured by Tissue Softness Analyzer, the data were combined with bending and tensile properties measured by the KOTITI Testing and Researching Institute and used for clustering and classification of textiles.

Tactile Sensation Analyzer (TSA) is a novel instrument which has a working principle similar to Tissue Softness Analyzer, and it is designed to evaluate hand of textile structures produced with various technologies including nonwovens, leather products, and a wide range of woven and knitted fabrics. TSA measures surface variations, deformation behavior, and recovery characteristics of fabrics via two simple measurement steps. Micro-surface variations (TS7) and macro-surface variations (TS750) of fabrics are measured using a unique sound analysis technique, meanwhile deformation (D), elasticity (E), hysteresis (H), and plasticity (P) properties are measured in the deformation test stage (Uren 2024). The device consists of a single desktop unit which implements all necessary measurements in succession, using the same test sample, and the test unit is connected to a computer where the test results are immediately displayed by the software.

The aim of this study was to evaluate tactile comfort of woven fabrics, knitted fabrics, and towels, and introduce the use of TSA to measure hand-related properties of fabrics with prominent structural differences. For this purpose, woven bed sheets, terry towels, and weft knitted casual knitwear fabrics were supplied from local and international manufacturers. Micro-surface variations (TS7), macro-surface variations (TS750), deformation (D), elasticity (E), hysteresis (H), and plasticity (P) of the fabrics were measured by TSA. Total hand (TH) scores of fabrics were determined by assessments of experts via sensory tests. In addition to the indicated evaluations, hand of fabrics was investigated by a conventional nozzle test equipment. Factors affecting hand of investigated fabric types were discussed, and correlation relations observed among TSA results, TH scores, and nozzle test results were presented.

Materials and methods

Materials

Fabrics were studied under three categories; woven bed sheets having various raw materials, settings, and yarn counts (Table 1), double-sided 3-pick terry towels with different unit weights, pile ratios and fiber contents (Table 1), and non-colored weft knitted cotton fabrics, produced with circular knitting technology (Table 2). As large number of samples may cause sensory overload during subjective evaluations, it was intended to maintain an optimum sample size for each category. The main objective when selecting materials for the study was to include fabrics with distinct differences in terms of hand, raw material, and structural parameters so that the samples in a particular category would cover the whole range of total hand scale.

Table 1. Structural properties of bed sheets and towels.

Category		Weave pattern	Material	Warp setting	Weft setting	Pile setting	Warp count	Weft count	Pile count	Unit weight	Thickness
			%	cm^-1	cm^-1	cm^-1	tex	tex	tex	g/m^2	mm
Bed sheets	B1	Plain	CO/PES (48/52)	24.0	15.0	-	20	30	-	97	0.325
	B2	Plain	CO	33.0	21.0	-	19	19	-	112	0.358
	B3	Plain	CO	33.0	24.0	-	20	18	-	118	0.336
	B4	Plain	CO/CLY (85/15)	45.0	33.0	-	15	16	-	130	0.299
	B5	Plain	CO	41.5	32.0	-	15	15	-	117	0.294
	B6	Sateen	CO/CLY (60/40)	78.0	47.0	-	10	10	-	128	0.301
	B7	Plain	CO/BB (30/70)	36.0	28.5	-	20	22	-	148	0.341
Towels	T1	3-pick terry	CO	13.0	20.5	13.0	2×29	43	2×30	516	3.87
	T2	3-pick terry	CO	14.0	17.0	14.0	2×29	36	38	441	2.91
	T3	3-pick terry	CO^m	14.0	17.5	14.0	2×29	37	45	474	5.12
	T4	3-pick terry^v	CO	14.0	22.0	14.0	2×30	36	30^f 36^b	428	2.87
	T5	3-pick terry	CO/PES (65/35)*	12.0	17.0	12.0	60	36	30	322	2.63
	T6	3-pick terry	CO/BB (60/40)**	12.0	19.5	12.0	2×30	36	37	501	3.69
	T7	3-pick terry	CO/BB (35/65)***	13.0	16.0	13.0	2×34	36	40	432	2.47

BB: bamboo-viscose, CLY: lyocell, CO: cotton, PES: polyester, ^mMicro Cotton (a trademarked term for fine natural luxurious cotton yarn woven into long loops for added absorbency and softness).

*Material of pile yarns: 100% cotton.

**Material of pile yarns: cotton/bamboo blend.

***Material of pile yarns: 100% bamboo.

^vterry towel with sheared face side (velvet effect).

Pile setting is given as the total number of face piles and back piles in unit fabric length.

^fface pile, ^bback pile.

Table 2. Structural properties of knitted fabrics.

	Knitting pattern	Material	Wale setting	Course setting	Yarn count	Unit weight	Thickness
		%	cm^-1	cm^-1	tex	g/m^2	mm
K1	2-thread fleece	CO	11.0	30.0	29^f 59^b	281	1.15
K2	Interlock	CO /EA (98/2)	21.2	24.0	20	284	1.21
K3	Single jersey	CO /EA (96/4)	15.0	23.0	20	219	0.92
K4	1×1 rib	CO	22.0	20.0	25	291	1.09
K5	1×1 rib	CO	21.0	20.0	20	216	1.05

CO: cotton, EA: elastane.

^fface (ground) yarn, ^bback (fleecy) yarn.

Sensory tests

Sensory evaluations were carried out by participation of 12 expert assessors (9 females and 3 males) aged between 27 and 49. Fabrics in different categories were evaluated in separate sessions. To ensure that all the fabrics were evaluated twice in each of several sessions, two samples (70 mm x 70 mm) were prepared for each fabric type. Separate sets of samples were prepared for all assessors.

Samples were presented to the assessors in a random positioning and order. The purpose was to increase the accuracy of sensory evaluations by preventing context, order, and position effects. During the assessments, panel members were asked to rank the samples using a five-point scale based on their tactile sensation. The scale points of total hand (TH) score were introduced as poor (1), fair (2), average (3), good (4), and excellent (5) tactile comfort. The evaluation process was unconstrained, which means that the assessors were free to evaluate the fabric hand according to the attributes they prioritize the most. To exclude bias caused by fabric appearance, assessments were carried out according to blind test requirements. No time limit was enforced. Evaluations were carried out as two replicates. Minimum duration between replicates was 10 days.

Tactile Sensation Analyzer

TSA is a multifunctional measuring instrument developed by the German company Emtec Electronic GmbH (Figure 1). The device characterizes surface properties and low-stress mechanical properties by sound analysis and deformation measurement. Symbols and descriptions of parameters measured by Tactile Sensation Analyzer are given in Table 3.

Figure 1. Measuring unit of Tactile Sensation Analyzer and its components.

Table 3. Parameters measured by Tactile Sensation Analyzer.

Symbol	Unit	Parameter	Description
TS7	-	Micro-surface variations	Lower value: lower micro-surface variations, higher smoothness
TS750	-	Macro-surface variations	Lower value: lower macro-surface variations, higher smoothness
D	mm/N	Deformation	Higher value: higher out-of-plane deformation, lower rigidity
E	mm/N	Elasticity	Higher value: higher elasticity, lower rigidity
H	J	Hysteresis	Lower value: lower energy generated during recovery
P	μm	Plasticity	Value closer to zero*: lower plastic (permanent) deformation

*In the current study, absolute values of plasticity were used.

Sound analysis

For sound analysis, the measuring head with the rotor moves down onto the sample and rotates on the surface, interacting with the fabric and protruding fibers (Figure 2a). This rotation motion happens at a load of 100 mN and it leads to two different vibrations. The vibration of the sample provides information about the macro-surface variations (TS750), while the vibration of the blades provides information about the micro-surface variations (TS7) (Figure 2b). The sound caused by these vibrations is recorded by two microphones, placed above and below the test sample.

Figure 2. Graphical representation of (a) rotation of the blades and (b) the sound spectrum recorded by TSA during sound analysis.

Deformation measurement

Deformation (D), elasticity (E), plasticity (P), and hysteresis (H) parameters are measured by TSA in the deformation test stage. Deformation measurement has three successive steps: the first cycle of deformation, recovery and the second cycle of deformation. In the beginning of the first cycle, blades of the measuring head contact with the fresh test sample with a load of 100 mN. Then the measuring head moves further down until a 600 mN load is reached, and the magnitude of out-of-plane deformation (D) is measured based on the vertical displacement of the measuring head (Figure 3a).

Figure 3. Graphical representation of (a) the vertical motion of the blades, and (b) deformation (D), elasticity (E), plasticity (P), and hysteresis (H) measured by TSA.

In the recovery step, the measuring head moves up, back to 100 mN load, and the recovery parameters – plasticity (P) and hysteresis (H) – are measured. Hysteresis is the energy generated during recovery and plasticity is the magnitude of permanent deformation (Figure 3b). For the second deformation cycle, the measuring head moves downwards until a 600 mN load is reached. The deformation recorded in this cycle is referred as elasticity (E). The main difference between parameters D and E is that D is measured with the fresh test sample and E is measured with the previously deformed sample, right after the recovery phase.

Test procedure

Samples with 120 mm x 120 mm dimensions were prepared for measurements. Mass per unit area and thickness of fabrics were entered to the software as two separate inputs. Mass per unit area was determined as described in ASTM D3776/D3776M–20 and thickness was measured under 5 gf/cm² pressure, using James Heal R&B Cloth Thickness Tester. All samples were conditioned and tested according to ASTM D1776/D1776M.

Nozzle test

Nozzle test is a conventional method where a fabric sample is fixed to a needle and pulled through a nozzle. The magnitude of pull-out forces recorded for each sample changes depending on several parameters such as bending rigidity, thickness, compressibility, and surface properties. In the current study, a nozzle construction having 24 mm diameter and 20 mm height, and a needle compatible with 10 N load cell were used. The construction was attached to Instron 4411 Tensile Tester working with constant rate of elongation principle. Circular samples having a diameter of 100 mm were prepared for bed sheets and knitted fabrics. Samples with a smaller diameter (70 mm) were prepared for towels. Nozzle tests were realized as five replicates. The test speed was 40 mm/min.

Load-displacement graph

It is commonly preferred to evaluate the hand of fabrics based on the maximum force recorded during nozzle test (Sular and Okur 2008; Uren and Okur 2019). However, several deformation stages can be observed during measurements. The load values recorded at these stages and the area under the graph can provide valuable information about fabric characteristics (El Mogahzy, Kilinc, and Hassan 2005; Pan and Yen 1992). In the current study, the load-displacement graphs obtained from the nozzle tests were examined in four deformation stages (Figure 4).

Figure 4. Deformation stages, pull-out forces and the sample geometry observed during nozzle test.

In the first stage of deformation, the sample touches the bottom of the nozzle and a hill (F1) in load-displacement graph is observed. In the second stage, the sample deforms into a cone shape and a peak (F2) is recorded. In the third stage, the continuous increase of the fabric volume inside the nozzle causes another hill (F3) and finally, in the last stage, the amount of fabric inside the nozzle decreases and a specific curve in load-displacement graph is observed. The load value which corresponds to the origin of this curve is recorded as F4. It must be noted that the displacement intervals of deformation stages and the profile of the load-displacement graph may change depending on the sample size and nozzle construction. Therefore, the ranges of deformation stages presented in Figure 4 were calculated for test parameters of the current study and for samples having a diameter of 100 mm. Considering the sample diameter, the displacement intervals of deformation stages were adjusted for towels by multiplying the displacement values with 0.7.

Normalization

To eliminate the bias caused by prominent structural parameters, nozzle test results were normalized. In the current study, effects of structural parameters such as thread count, stitch density, and thickness on nozzle test results were investigated based on correlation coefficients calculated between results of nozzle test and sensory evaluations. Accordingly, for normalization, pull-out forces recorded for bed sheets, knitted fabrics, and towels were divided by thread count, stitch density and fabric thickness, respectively.

Statistical analysis

Performance of assessors on the descriptive sensory panel was evaluated based on Kendall’s coefficient of concordance (Kendall’s W) values. Significance of differences between paired variables was investigated using Wilcoxon signed rank test. One sample Kolmogorov Smirnov test was used to determine normality and uniformity of variables. The results were evaluated at 95% confidence level. Relations between variables were investigated by Spearman’s rank-order correlation analysis. The significant relations observed at 90%, 95%, and 99% confidence levels were reported.

Results

Sensory evaluation results

Performance of sensory panel

Before estimating the accuracy of a test method based on sensory evaluation results, it is important to examine the agreement and reliability of panel members. In the current study, the performance of assessors on the sensory panel was investigated based on concordance coefficients (Kendall’s W). The high concordance values associated with inter-rater agreement assured that the assessors applied essentially the same standard when assessing tactile comfort of the samples (W ≥ 0.848). The Kendall’s W values calculated for results collected over multiple sessions were between 0.731 and 1.00 with an average of 0.923, which indicated an excellent level of intra-rater reliability (Table 4).

Table 4. Kendall’s coefficient of concordance values calculated for inter-rater agreement and intra-rater reliability.

	Assessor	Bed sheets	Towels	Knitted fabrics
Intra-rater reliability	A1	0.995	1.000	0.961
	A2	0.983	0.895	0.912
	A3	1.000	0.977	0.868
	A4	0.911	0.982	0.944
	A5	1.000	0.820	0.910
	A6	0.981	0.995	0.912
	A7	0.819	0.982	0.950
	A8	0.959	0.945	0.833
	A9	0.936	0.982	0.987
	A10	0.810	0.914	0.957
	A11	0.973	0.907	0.731
	A12	0.857	0.889	0.750
Inter-rater agreement		0.852	0.848	0.888

Agreement among assessors (inter-rater agreement) was given as Kendall’s W values, N = 12.

Agreement between two replicate trials (intra-rater reliability) was given as Kendall’s W values calculated for each assessor, N = 2.

Total hand scores

It was observed that TH scores of samples were covering a large portion of the 5-point scale (Figure 5) and the distributions of TH scores were statistically normal and uniform for each category (p ≥ .199).

Figure 5. Distribution of total hand scores of bed sheets (B), towels (T) and knitted fabrics (K).

Results of Tactile Sensation Analyzer

Surface properties

TSA measures surface properties in sound analysis stage. In this stage, sound intensity between 0–10000 Hz area was recorded and the intensities at two specified peaks (TS7 and TS750) were recorded (Figure 2). Sound spectra of investigated fabrics were summarized using average sound intensity values of samples within the same category and presented in Figure 6a.

Figure 6. Surface properties of bed sheets (B), towels (T), and knitted fabrics (K); (a) sound spectra recorded by TSA, and (b) micro and macro-surface variations and their relations with total hand scores.

The intensity of the peak before 1000 Hz (TS750 peak) indicates the magnitude of macro-surface variations. As can be seen in Figure 6a, the highest TS750 peak was recorded for towels. The piles of towels create great variations on the fabric surface, and the TS750 values recorded for towels were in accordance with this information. It was observed that the towel sample with the lowest TH score (T1) has the highest macro-surface variations. The towel samples made of cotton/bamboo and cotton/PES blends (T5, T6, T7) exhibited the lowest macro-surface variations and they were rated with the highest TH scores (Figure 6b).

Weft knitted samples evaluated in the current study were non-colored cotton fabrics which are commonly used in casual knitwear products. The loop structure of investigated knitted fabrics creates smaller macro-surface variations when compared to towels but larger macro-surface variations than woven bed sheets. Certifying the aforementioned observations, the average TS750 value of knitted fabrics was lower than towels and higher than bed sheets (Figure 6a). It was recorded that face and back sides of two-thread fleece knitted sample (K1) have differentsurface structures (Figure 6b) due to the special knitting technique which enables finer ground yarns to form a regular plain weave structure on the face side, meanwhile coarser fleecy yarns form floats on the back side of the fabric. Sample K1 also exhibited the highest macro-surface variations among knitted samples, and it was rated with the lowest TH score.

The TS7 peak recorded around 6500 Hz indicates the magnitude of micro-surface variations. As can be seen in Figure 6a, TS7 peak of bed sheets was higher than towels and knitted fabrics. Unlike towels and knitted fabrics, for textiles such as bed sheets which do not exhibit large macro-surface variations, the magnitude of micro-surface variations may become more relevant in terms of fabric hand. In this context, comparing micro-surface variations of this type of fabrics may provide a better insight about tactile characteristics. It was detected that back side of the cotton/lyocell blend sample with sateen weave pattern (B6) and face side of the sample made of cotton/bamboo blend (B7) have the lowest micro-surface variations among all bed sheet samples and they were rated with the highest TH scores (Figure 6b).

While hand of fabrics might be more or less related to a particular fabric property, it is expected to observe significant correlation relations between sensory evaluations and objective measurements. In this context, relevance of TH scores and TSA results was investigated by correlation analysis. When discussing surface properties, it is essential to indicate which side of the sample was subjected to surface characterization. In the current study, both sides of the samples were tested. Investigations indicated that there is a significant difference between surface characteristics of face and back sides of the fabrics (p ≤ .043). Therefore, in the current study, correlation relations between TH scores and surface parameters (TS7 and TS750) were investigated for face and back sides separately (Table 5).

Table 5. Spearman’s correlation coefficients (r_s) calculated among total hand scores, surface characteristics, low-stress mechanical properties and normalized nozzle test results of bed sheets, towels and knitted fabrics.

Category			Total hand score	Pulling force^n
				F1	F2	F3	F4	Fmax
Bed sheets	Total hand score		/	−0.893***	−0.929***	−0.964***	−0.857**	−0.929***
	Micro-surface variations (TS7)	Average	−0.893***	0.714*	0.964***	0.929***	0.929***	0.964***
		Face	−0.821**	0.786**	0.750*	0.714*	0.714*	0.750*
		Back	−0.929***	0.821**	1.00***	0.964***	0.964***	1.00***
	Macro-surface variations (TS750)	Average	−0.786**	0.929***	.	0.714*	.	.
		Face	−0.857**	0.893***	0.679*	0.750*	.	0.679*
		Back	−0.714*	0.857**	0.714*	0.786**	0.679*	0.714*
	Elasticity (E)	Average	0.750*	−0.786**	.	.	.	.
	Hysteresis (H)	Average	−0.893***	0.786**	0.821**	0.857**	0.714*	0.821**
	Plasticity (P)	Average	−0.786**	.	0.750*	0.821**	.	0.750*
Towels	Total hand score		/	−0.714*	/	−0.786**	.	−0.786**
	Micro-surface variations (TS7)	Average	−0.750*	.	/	.	.	.
		Face	−0.679*	.	/	.	.	.
		Back	−0.786**	0.786**	/	0.821**	0.714*	0.821**
	Macro-surface variations (TS750)	Average	−0.893***	0.821**	/	0.786**	.	0.786**
		Face	−0.857**	0.857**	/	0.857**	0.714*	0.857**
		Back	−0.857**	0.857**	/	0.750*	.	0.750*
	Deformation (D)	Average	0.857**	−0.857**	/	−0.750*	.	−0.750*
	Elasticity (E)	Average	0.857**	−0.857**	/	−0.750*	.	−0.750*
	Plasticity (P)	Average	0.786**	−0.929***	/	−0.893***	−0.786**	−0.893***
Knitted fabrics	Total hand score		/	−0.900**	−1.00***	−1.00***	−0.900**	−1.00***
	Macro-surface variations (TS750)	Average	−1.00***	0.900**	1.00***	1.00***	0.900**	1.00***
		Face	−1.00***	0.900**	1.00***	1.00***	0.900**	1.00***
		Back	−1.00***	0.900**	1.00***	1.00***	0.900**	1.00***
	Plasticity (P)	Average	.	−0.900**	.	.	.	.

Correlation is significant at the 0.10 level (*), 0.05 level (**) and 0.01 level (***), (2-tailed).

ⁿNormalized pulling forces: Fmax is the maximum pulling force and F1, F2, F3, F4 are pulling forces recorded during the 1st, 2nd, 3rd and 4th stages of nozzle test.

For normalization, nozzle test results of bed sheets, towels and knitted fabrics were divided by thread count, thickness and stitch density, respectively.

It was observed that fabrics with lower surface variations were rated with higher TH scores in general (Figure 6b). In accordance with this observation, micro and macro-surface variations of studied fabrics were found out to be inversely proportional to TH scores (Table 5). The strength and significance of these correlation relations were different depending on fabric category. As previously suggested, when working with fabrics which have large variations on their surfaces, comparing macro-surface variations may provide a more relevant data regarding tactile characteristics. On the other hand, for fabric types with relatively smoother surfaces, investigating micro-surface variations may establish a better insight on fabric hand. Statistical calculations confirmed that hand of towels and knitted fabrics was greatly affected by macro-surface variations (p ≤ .014) meanwhile TH scores of bed sheets exhibited the highest correlation relations with micro-surface variations (p ≤ .023) (Table 5). It was also recorded that hand of towels was related to micro-surface variations and hand of bed sheets was related to macro-surface variations, however significance of these correlation relations was relatively low (p ≤ .094) (Table 5).

Low-stress mechanical properties

Low-stress mechanical properties of fabrics were measured by TSA at deformation test stage. The first parameter measured in this stage is out-of-plane deformation (D), which indicates the magnitude of deformation observed on the sample when the load applied on the sample was increased from 100 mN to 600 mN (Figure 3). In the second cycle of deformation measurement, the same load was applied to the same sample for the second time and the deformation observed in this cycle was recorded as elasticity (E). Besides deformation and elasticity, recovery characteristics of fabrics were measured by TSA as well. The energy generated during recovery was measured and referred as hysteresis (H) and the level of permanent deformation was recorded as plasticity (P). Unlike surface properties, in the current study, no significant difference between face and back side measurements was detected for low-stress mechanical properties (p ≥ .082). Accordingly, for parameters D, E, P and H, only the average values of face and back side measurements were used in graphs and statistical investigations.

Bed sheets need to be designed to meet specific mechanical and comfort requirements (Sundaresan et al. 2016). Even though the ability of a fabric to deform under a low load is an essential element of tactile comfort to a certain extent, a high permanent deformation (P) may not be a desirable quality for a bed sheet. It was previously reported that deformation and elasticity of conventional woven fabrics were primarily related to bending rigidity, extensibility, and resistance to repeated shear, meanwhile recovery characteristics (P and H) were stated to be related to bending length and extensibility recorded under 5 N/m load (Uren 2024). In the current study, it was observed that deformation and elasticity of bed sheets were noticeably lower than knitted fabrics and towels (≤2.35 mm/N). In fact, the lowest D and E values were recorded for the bed sheet made of cotton/PES blend (B1) (Figure 7). It was observed that the sample with cotton/bamboo blend (B7) has the highest D and E values among all bed sheet samples.

Figure 7. Low-stress mechanical properties of bed sheets (B), towels (T) and knitted fabrics (K) measured by TSA, and their relations with total hand scores.

Results of correlation analysis indicated that bed sheets with higher elasticity and better recovery characteristics – lower hysteresis and plasticity – were rated with higher TH scores (p ≤ .052) and hysteresis was the most prominent low-stress mechanical property for determining hand of bed sheets (p = .007) (Table 5). The inverse proportion detected between TH score and plasticity of bed sheets indicated that assessors have an aversion to bed sheets with higher permanent deformation. Sample B1 which has the highest plasticity and hysteresis results was rated with the lowest TH score (1.96). Samples B6 and B7 exhibited the highest TH scores (≥4.58) and the lowest plasticity and hysteresis values (≤105 µm and ≤ 48 J, respectively).

It is widely recognized that weft knitted fabrics have significantly higher extensibilities when compared to conventional woven structures. In addition to that, using elastic fibers can greatly improve the stretch and recovery capability of fabrics (Luo et al. 2011). In the current study, the highest D and E values (≥4.29 mm/N) were recorded for knitted fabrics which were produced with cotton/elastane blends (K2 and K3) (Figure 7). In this respect, the magnitudes of out-of-plane deformations recorded at the first and the second deformation cycles (D and E) were assumed to be mostly related to the high extensibility of knitted fabrics. For materials with a good recovery characteristic and high extensibility, it is expected to observe similar deformation behaviors in repeating cycles. Supporting the mentioned assumption, the difference between D and E values of knitted samples was less than 7.4%. It was observed that two-thread fleece knitted sample (K1) has significantly lower deformation and elasticity and a relatively high hysteresis, and this sample has the lowest TH score when compared to the other knitted samples (1.44) (Figure 7). Sample K5 exhibited the highest permanent deformation (326 µm) and it was rated with the highest TH score (4.67). Even though some relations were observed between TH scores and low-stress mechanical properties of knitted fabrics (Figure 7), these relations were not significant from a statistical perspective (p ≥ .100).

Similar to knitted fabrics, deformation (D) results of towels were quite high (between 1.89 and 3.85 mm/N). The maximum deformation observed on towel samples under a load increase of 500 mN was calculated to be between 0.95 and 1.93 mm, which was lower than the initial thickness of the fresh samples (≥2.47 mm). This result indicated that the deformation observed on towel samples is rather a compression deformation than an extension or a bending deformation occurred in out-of-plane direction. Consequently, the D values recorded for towels were assumed to be related to the compressibility of pile yarns, not the overall extensibility of the fabric. Calculations also indicated that, for towels, parameter E represents the magnitude of repeated compression deformation. Supporting this assumption, the second cycle deformation results (E values) of towels were observed to be noticeably lower than D (between 1.63 and 2.63 mm/N). In fact, unlike knitted fabrics, a great difference between D and E values was recorded for towels (between 14% and 32%) (Figure 7).

In the current study, towels exhibited the highest recovery energy and the largest permanent deformation (≥71.6 J and ≥170 µm, respectively) (Figure 7). The high levels of permanent deformation (P) recorded for towels certified that the piles on the surface of towel samples were already in a compressed state after the first cycle of deformation (Figure 7). It was previously reported that using single or plied yarns has a significant effect on hand of towels (Udurgucu, Sancar Besen, and Serafettinoglu 2024). In accordance with the previous literature, it was recorded that the sample produced with two-plied pile yarns (T1) has significantly lower deformation, plasticity and hysteresis results among all towel samples and it was rated with the lowest TH score (1.21). The samples made of single plied cotton and bamboo pile yarns (T5, T6, T7) exhibited the highest deformation and plasticity values and they have the highest TH scores (≥4.19) (Figure 7). Statistical calculations certified that TH scores of towels were strongly correlated with and directly proportional to out-of-plane deformation (D), elasticity (E) and plasticity (P) (p ≤ .036) (Table 5). Results of the current study revealed that P values measured with TSA were related to compressibility of pile yarns. Consequently, the direct proportion recorded between TH scores and plasticity of towels indicated that towels with less rigid piles were rated with higher TH scores.

Nozzle test results

To establish the relations between nozzle test and sensory evaluation results, correlation coefficients between TH scores and normalized pull-out forces were calculated (Table 5). It was observed that sensory evaluation results were strongly related with and inversely proportional to the normalized load values of bed sheets and knitted fabrics. Similarly, F1, F3, and Fmax values of towels were strongly correlated with TH scores. As suggested by the literature, Fmax values provided high and significant correlation relations with sensory evaluations in general (Sular and Okur 2008; Uren and Okur 2019). It must be noted that, during the nozzle test, towel samples were deformed in a gradual manner. Therefore, no peak value at the second stage of nozzle test (F2) was recorded for towels.

The findings indicated that normalized pull-out forces (F1, F2, F3, F4, and Fmax) were directly proportional to the magnitude of surface variations (Table 5). Macro-surface variations (TS750) of knitted fabrics and towels, and micro-surface variations (TS7) of bed sheets were recorded to be highly and significantly correlated with most of the pull-out forces. Meanwhile, macro-surface variations of bed sheets were found out to be highly correlated with F1 and moderately correlated with F2, F3, F4, and Fmax.

The pull-out forces recorded during several stages of nozzle test were inversely proportional to out-of-plane deformation (D) and elasticity (E). Results of correlation analysis indicated that the significance and direction of correlation relations calculated between recovery characteristics (H and P) and pull-out forces change depending on the fabric type.

Peak value of the first hill (F1) in load-displacement graph of nozzle test was recorded when the sample touches the bottom of the nozzle for the very first time. Therefore, F1 values indicate a sample’s initial reaction to folding. Results of knitted fabrics indicated that F1 was inversely proportional to plasticity of knitted fabrics (p = .037), and it was the only pull-out force which was significantly correlated to TSA results (Table 5).

Bed sheets with higher elasticity (E) and better recovery (lower plasticity and hysteresis) exhibited lower levels of resistance when being pulled through the nozzle. Hysteresis (H) of bed sheets was significantly correlated with all pull-out forces, whereas elasticity (E) was correlated with F1, and plasticity (P) was correlated with F2, F3 and Fmax (Table 5).

As formerly mentioned, deformation (D), elasticity (E), and plasticity (P) results of towels represent different aspects of compressibility of the pile yarns. In the current study, load values recorded when the towel sample was pulled through the nozzle were detected to be inversely proportional to D, E and P. Plasticity of towels was found out to be highly correlated with all pull-out forces. Deformation and elasticity were highly correlated with F1 and moderately correlated with F3 and Fmax (Table 5).

Discussion

Tactile comfort is an essential element of textile products which are in a regular contact with human skin. Cellulose-based fibers are commonly preferred in production of next-to-skin clothing items and several home textiles because of their improved comfort properties. In the current study total hand (TH) scores of cellulose-based textile products – woven bed sheets, terry towels and weft knitted fabrics – were determined by sensory tests. Micro-surface variations (TS7), macro-surface variations (TS750), deformation (D), elasticity (E), hysteresis (H), and plasticity (P) of the fabrics were measured using Tactile Sensation Analyzer (TSA). Hand of fabrics was also investigated by a nozzle test equipment and the pull-out forces recorded at certain regions of load-displacement graph (F1, F2, F3, F4, and Fmax) were normalized and reported.

Hand of bed sheets was found out to be significantly correlated with most of the surface and low-stress mechanical properties measured by TSA, where micro-surface variations and hysteresis parameters showed the highest correlation relations. It was detected that bed sheets produced with cotton/bamboo (30/70) and cotton/lyocell (60/40) blend fine yarns have smoother surfaces, higher elasticity values and better recovery characteristics, and these fabrics were rated with higher TH scores.

Results of the current study pointed that surface and low-stress mechanical properties of towels were strongly correlated with TH scores – except hysteresis. It was concluded that deformation, elasticity, and plasticity of towels measured using TSA represent compressibility of the pile yarns, and these parameters were directly proportional to the TH scores. Findings indicated that pile yarns made of cotton and bamboo provided the most desirable tactile comfort for towels. Investigations certified that tactile comfort of towels was strongly affected by macro-surface variations and the compressibility of pile yarns, and moderately affected by the magnitude of micro-surface variations.

TH scores of knitted fabrics were significantly correlated with and inversely proportional to macro-surface variations. Some correlation relations were observed between TH scores and micro-surface variations, deformation, elasticity, and hysteresis of knitted fabrics as well, but these relations were statistically not significant.

The nozzle test equipment used in the current study was observed to be compatible with the investigated fabric types. However, the raw data obtained by this test method were not correlated with sensory evaluation results unless they were normalized. Therefore, pull-out forces recorded for bed sheets, knitted fabrics and towels were divided by thread count, stitch density and fabric thickness for normalization, respectively. It was observed that normalized nozzle test results were significantly correlated with TH scores, surface characteristic, and low-stress mechanical properties.

When working with a novel test instrument, it is essential to investigate the reliability of the test results and consider the possible limitations and constraints in scientific terms. TSA is a newly developed test instrument which uses a unique sound analysis technique to identify micro and macro-surface variations and measures low-stress mechanical properties during the deformation test. Unlike well-known measurement systems which consist of a number of separate test modules, TSA consists of a single test unit, and instead of measuring each parameter separately for each test direction (warp, weft, diagonal etc.) in a conventional manner, the device determines surface variations in a circular area, and deformation tests are carried out by applying a load on the test sample from multiple axes to create an out-of-plane deformation. The device also uses the same test sample for both sound analysis and deformation measurements. Consequently, the test duration is short, and the required specimen size is quite small. The highly significant correlation relations recorded between TSA results and TH scores verified that TSA can be used to determine hand of bed sheets, towels, and weft knitted fabrics. Consequently, it was concluded that TSA may meet the increasing demand for instruments for objective measurement of tactile sensory properties of fabrics produced with different technologies and provide several benefits for manufacturers and researchers.

Conclusions

Considering the time and labor factors, it is important to develop an objective test method which would provide results consistent with sensory evaluations. Textiles may have very different structural and mechanical properties. However, most of the conventional test instruments and methods are applicable to woven structures. The aim of this study was to investigate hand of cellulose-based fabrics used as bed sheets, terry towels and casual knitwear products. The components of fabric hand – surface characteristics and low-stress mechanical properties – were determined using “Tactile Sensation Analyser” (TSA) which is a test instrument designed for measuring a wide range of textile structures.

Surface parameters measured by the sound analysis technique of TSA were found out to be strongly related to the perceived tactile comfort, and the TH scores determined by sensory evaluations were detected to be inversely proportional to the magnitude of surface variations. Hand of bed sheets was highly correlated with micro-surface variations and moderately correlated with macro-surface variations. TH scores of towels and knitted fabrics were mainly related to macro-surface variations. These findings indicated that macro-surface variations may provide a more valuable data when evaluating hand of fabrics with irregular surface characteristics, meanwhile micro-surface variations may be a more determinant attribute for hand of fabrics with relatively smoother surfaces.

Results certified that deformation and recovery properties of bed sheets and towels have a significant effect on their tactile comfort. It was observed that bed sheets with higher elasticity and better recovery characteristics were rated with higher TH scores. Deformation, elasticity, and plasticity of towels measured by TSA were detected to represent the compressibility of the pile yarns, and samples with more compressible pile yarns were rated with higher TH scores. Results of the study also revealed that producing towels and bed sheets with cotton/bamboo blends provided a significant improvement in tactile comfort.

Hand of fabrics was also investigated by a nozzle test equipment and the pull-out forces (F1, F2, F3, F4 and Fmax) recorded during nozzle test were normalized by thread count, stitch density or fabric thickness. Nozzle test results exhibited significantly high correlation relations with TH scores after the suggested normalization process. Similarly, normalized pull-out forces were significantly correlated with surface and mechanical properties of fabrics.

TSA was observed to be a user-friendly instrument which is capable of measuring surface and low-stress mechanical properties of various types of fabrics with a single device and within a very short duration. Surface variations of the investigated fabric types determined by the sound analysis technique of TSA and the low-stress mechanical properties measured in out-of-plane deformation state were detected to be significantly correlated with TH scores. Based on these observations, it was concluded that hand-related properties of textiles with distinct structural differences can be determined using TSA, and the proposed device can be a good alternative to integrated measurement systems and test equipment. It is believed that the test results presented in the study could provide valuable data for both researchers and commercial entities.

Highlights

• Hand of bed sheets was highly related to surface, elasticity and recovery properties.

• Hand of towels was greatly affected by surface characteristics and compressibility of pile yarns.

• Hand of knitted fabrics was significantly correlated with macro-surface variations.

• Results of Tactile Sensation Analyzer (TSA) were validated by sensory evaluations.

Acknowledgments

The authors would like to thank to the assessors participated in sensory evaluations. We also would like to thank to Emtec Electronic GmbH for their cooperation in this study.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.