Prediction of Psychological Comfort Properties of 100% Cotton Plain Woven Fabrics made from Yarns with Diff

The psychological satisfaction of the textile customer is the fi rst criteria used to evaluate clothing and a lack of aesthetics, while fashionability and physical appearance contribute to the psychological discomfort of users. Either inherently or due to processing, cotton cloths demonstrate diff erent psychological comfort behaviours. Manufacturers must therefore produce fabrics with optimum psychological comfort parameters. The objective of this research was to study the eff ect of cotton yarn parameters on the psychological comfort properties of woven fabrics. Four woven fabrics were produced from cotton yarns with diff erent yarn twists, yarn counts, strengths and yarn elongations. Psychological comfort parameters such as wrinkle, drape, crease, bending modules and fl exural rigidity were measured and analysed in accordance with the ES ISO 9867, ISO 9073-9, ES ISO 2313 and ASTM D1388-18 standards, respectively. Multiple regression equations were developed to predict the comfort properties of clothing in relation to yarn parameters. A statistical analysis showed that the wrinkle recovery and drapeability of fabrics were signifi cantly aff ected by yarn twist, count and tenacity, and the elongation of yarns. However, yarn twist, count, tenacity and elongation had an insignifi cant eff ect on the crease recovery of woven samples at an F-value of 3.546 and a P-value of 0.069. The stiff ness properties of the fabrics such as fl exural rigidity and bending modules also showed insignifi cant diff erence between samples at F = 38487.969, P = 0.057 and F = 25.506, P = 0.055 respectively. A multiple regression analysis showed a positive correlation between yarn parameters (factors) and response, with Adj. R2 of 0.0998, Adj. R2 of 0.975 and Adj. R2 of 1 for crease recovery, wrinkle recovery and drape coeffi cient, respectively. The equations developed are helpful to fabric manufacturers in sourcing yarns with specifi c parameters to produce the desired comfort level in a fabric.


Introduction
Psychological comfort has received attention from manufacturers in recent years due to consumers' demands for aesthetic value, fashionability and good clothing appearance, as factors considered in the purchase process. To produce a comfortable garment, comfort properties should be considered during manufacturing and garment design [1]. Clothing with poor wrinkle, low drapeability and crease recovery properties will decrease wearers' psychological satisfaction. Natural fi bres have better comfort properties than synthetic fi bres. Even though they have several advantages, they also have some disadvantages, such as the quick wrinkling of cotton fabric during wear [2]. Yarn parameters are very important in producing suitable fabric with optimised performance for a specifi c end use. Several researchers have shown that yarn properties aff ect the properties of clothing. Gong studied cotton fabrics using fi ve control factors in order to examine their eff ects on yarn cross-sectional shape changes along the yarn path. Th e factors studied were fi bre type, yarn linear density and twist factor, and warp and weft cover factors. Th e study focused on the cross-sectional structure and its infl uence on woven fabric [3]. Pattanayak and Luximo pointed out that there are many factors that infl uence the aesthetic appearance of a fabric and the outstanding eff ect on the actual beauty of the cloth [4]. Drape is one of the critical factors aff ecting psychological comfort. Th e drape or drapeability of a fabric refers to the manner in which the fabric falls, shapes, gathers or fl ows with gravity on a model form or on a human body. Fabric drape has attracted the attention of many researchers in recent years because of the attempt to create a clothing CAD system by introducing fabric material properties in which drape is the key element [5]. Pattanayak and Luximo stated that fabric drape is an important parameter for the selection and development of textile materials for apparel industries. Predictions of the drape property of cotton woven fabrics were developed using multiple regression methods [4]. Another researcher investigated the improvement of the crease recovery of cotton fabrics. Reactive dyed fabrics were treated with dimethylol dihydroxy ethylene urea (DMDHEU) resin in order to improve their crease recovery characteristics. Two types of treatments were carried out: conventional thermal curing and gamma irradiation. Finally, the eff ect of treatments on crease recovery, and the mechanical and thermal properties of fabrics were studied. Th ey found that the fi nishing of cotton fabrics with gamma irradiation demonstrated better crease recovery [6]. Th e effect of yarn twist has also been studied. It was observed that the crease recovery of cotton fabrics decreases as yarn twist increases [7]. Similarly, Liu et al. investigated the impact of mechanical action on the wrinkling of cotton fabrics in a drum washer [8]. Th ey observed that the spinning speed and wash load were the main factors infl uencing the smoothness of cotton fabrics. As the wash load increased, the free motion region decreased and the ratio of the passive motion region increased, resulting in the severe wrinkling of cotton fabric. Hala [9] studied the eff ect of yarn twist direction on the drapability of fabrics and observed that twist direction does indeed have an eff ect on drape property. Th e bending and drape properties of woven fabrics, and the eff ect of weft density, weft yarn count and warp tension on these properties were also investigated by Süle [10]. It was reported that woven fabrics with thicker weft yarns and higher weft densities had a higher bending rigidity. In addition, the bending rigidities of the fabrics in the warp directions increased as warp tension increased. King and Johnston [11] studied the eff ect of stiff ness, shear, extensibility, thickness and density on the drape coeffi cient of fabrics. A multiple regressions analysis equation was developed to determine drape coeffi cient based on stiff ness, shear, extensibility and density. Th ey concluded that the drapeability characteristics of a fabric are aff ected, to some extent, by fi bre stiff ness [12]. Multiple regression models were built based on factors such as bending, shear, tensile, compression and aerial density. It was found that tensile and compression factors have little eff ect on the drape property of fabrics. It was observed that bending, shear and aerial density aff ect drape characteristics [2,4]. Th e eff ect of yarn parameters on the mechanical properties (at low-stress) of woven fabrics, tensile strength, compression, bending and shear property were studied by the authors of this work [13]. Th ough many researchers have studied the eff ect of yarn parameters such as count and twist on tactile comfort, and on drape, tensile and stiff ness properties, most of the studies were done at the grey fabric level. Th e eff ect of yarn properties on fabric wrinkle resistance have not been thoroughly addressed. The eff ects of yarn parameters such as yarn count, twist, strength and elongation on the psychological comfort properties of half-bleached woven fabrics were studied in this work. A multiple regression analysis was carried out to predict stiff ness, wrinkle recovery, crease recovery and the drape properties of fabrics.

Materials
Four types of cotton yarns were manufactured by Bahir Dar Textile S.C. using a C-60 IDF (Integrated Draw frame) machine and a rotor spinning system (RIETER-R 923). Th e yarns were selected randomly from daily production when the count and twist were changed by spinners on the same machine. All yarns samples were produced from the same fi bre mix with a micronaire value of 4.02, maturity of 0.85, upper half mean length (UHML) of 30.16 mm, uniformity index (UI) of 83.8%, short fi bre content of 6.9, strength of 29.8 g/tex, elongation of 7.6% and trash grade three. Trash grade is known as TrGrd in USTER HVI 1000 machine and this value is measured by the trash module and consist of 1 up to 4 alphanumeric characters. Yarn twist, yarn count, strength and elongation were evaluated according to ASTM D1422, ES ISO 2060 and ES ISO 2062 testing standard methods, respectively, and are shown in Table 1. A Statimat Me+ instrument was used to evaluate yarn strength and elongation. Table 2 shows the actual thread density and thickness of the fabrics, as well as nominal yarn count and twist. All woven fabrics were produced on the same air jet loom (PICANOL-OMP 800-2-P) at a speed of 350 min -1 (350 rpm) by Bahir Dar Textile S.C. Machine speed, warp yarn property, thread density (warp = 20 cm -1 and weft 18 cm -1 ) and fabric structure (plain) were kept constant (see Tables 1 and 2) during fabric production. Th e only variation was the use of diff erent weft yarns. Th e produced fabric samples were treated using a halfbleach combined treatment system on a Jigger machine. Th e fabric and water solution were prepared at a material liquor ratio (MLR) of 1:5, H 2 O 2 4% of fabric weight, NaOH 3% of fabric weight and Na 2 SiO 3 2% of fabric weight, and a wetting agent of 0.5% of fabric weight. Each sample was treated at a temperature of 95 ºC for 1.5 hours, with a machine working speed of 40 m/minute.

Test methods
Specimens were conditioned at a relatively humidity of (65 ± 2)% and a temperature of (20 ± 1) °C before each test.

Structural properties
Th e structural parameters of the developed fabrics such as thread density and fabric weight were evaluated according to ESISO 7211-2 and ES ISO 3801 test methods, respectively. Th e nominal thread density (set at loom) of the fabric and the actual thread density (aft er manufacturing) are presented in Table 2.

Stiff ness test
Fabric stiff ness is the resistance of a fabric to bending. Th is test measures the bending stiff ness of a fabric by allowing a narrow strip of the fabric to bend to a fi xed angle under its own weight. Th e length of the fabric required to bend to this angle is measured and is known as the bending length.
A cantilever stiff ness test system is oft en used as a measure of a fabric's stiff ness, as it is an easy test to perform. During testing, a horizontal strip of fabric is clamped at one end and the rest of the strip allowed to hang under its own weight as shown in Figure 1.

Figure 1: cantilever fabric stiff ness test method L -length of projecting fabric, Ѳ -angle to which fabric bends, M -fabric mass per unit area
Test specimens measuring 25 mm in width and 200 mm in length were cut in a lengthwise direction.
Five tests were carried out on each sample in the warp and weft direction according to the ASTM D1388-18 standard. A fabric stiff ness analysis, including parameters such as fl exural rigidity, bending modulus and bending length, was also performed. Th e bending length was measured, and the bending modulus and fl exural rigidity calculated according to equations 1 and 2: Flexural rigidity = MC 3 (1), where M is the weight of a fabric in g/m 2 , C is the bending length, G is fl exural rigidity and T is fabric thickness.

Drapeability test
Drape is the degree of fabric deformation when it is allowed to hang freely under its own weight. It is quantitatively expressed as a drape coeffi cient [11]. Th e drape coeffi cient is defi ned as the percentage of the area of the annular ring of fabric obtained by vertically projecting the shadow of the draped specimen. Th is test is used as an indication of garment appearance properties when a fabric orients itself into folds in more than one plane under its own weight. In this study, 30 cm diameter circles samples were used for testing on a Cusick drape tester in order to assess the drape of the fabric. Five tests were carried out on each sample in accordance with the ISO 9073-9 test method. Th e samples were positioned over a horizontally placed circular rigid disk of 18 centimetres in diameter. Th e fabric was deformed into a series of folds around the disk. Th e paper ring containing the shadow image of the draped confi guration represents the weight (W 1 ). Th e shadow image cut from the paper ring is weighed and marked as W 2 . Finally, the drape coeffi cient (DC) is calculated using equation 3: where W 2 is the mass of the shaded area and W 1 is the total mass of the paper ring.

Wrinkle property
Cotton fabric wrinkles easily and is therefore prone to shrinking, which decreases the wearer's psychological acceptance. A wrinkle may be explained as a rhytide, fold, ridge or crease in a cloth or garment. To test the wrinkling behaviour of the samples, their appearance was evaluated aft er crushing. Th e fabric sample was crushed and maintained at a specifi ed pressure and time under standard atmospheric conditions in a wrinkle tester. Th e sample was then removed and its appearance visually compared to a reference sample and rated according to the ES ISO 9867 tests standard.

Crease recovery
Creasing of a fabric during wear has a signifi cant infl uence on the wearer's psychological satisfaction an on viewer acceptance. Th us, all clothing must have good crease recovery properties. Crease recovery is the ability of a fabric to return from a collapsed deformed state. In this investigation, crease recovery was measured quantitatively in terms of the crease recovery angle. Textiles used in clothing must have the ability to crease and fold in order to conform to body shapes, and thus ensure improved comfort during wear. To retain their appearance, however, they must be able to eliminate creases that occur in wear and laundering. When a fi bre bends, cross-links may break and be formed in a new position, restricting recovery. Otherwise, they will merely stretch and recover when the load is removed. Th e M003A Shirley crease recovery tester was used for evaluation purposes. For the crease recovery test, 50 mm × 25 mm rectangular specimens were conditioned at a relative humidity of 65 ± 2% and a temperature of 20 ± 1 °C for 24 hours then folded for fi ve minutes under a 2-kg load as per the ES ISO 2313 test standard method. Th is creasing load was then removed and the specimen allowed to recover for another fi ve minutes in the crease recovery tester and the crease recovery angle recorded.

Structural properties
As observed in Table 2, the samples had the same fabric structure, thread density and loom settings. Nominal and actual thread densities were slightly diff erent, as shown in Table 2. Table 3 shows that fabric stiff ness, expressed as fl exural rigidity and bending modulus, did not change signifi cantly. Th e eff ects of yarn count, yarn twist, strength and elongation on woven fabrics were insignifi cant at an F-value of 38487.969 and Sig-value (Pvalue) of 0.057 on fl exural rigidity, and an F-value of   Table 2 were not significantly diff erent. Fabric thickness infl uences bending resistance, or rigidity, and may vary signifi cantly, depending on the structure and texture, while mass per surface area, or GSM, in itself is not an expressive property concerning fabric fl exibility in the absence of information on fabric thickness [14].

Drapeability of fabrics
As evident in Table 5, the produced woven fabrics showed a high signifi cant change in drape coefficient with an F-value of 113.610 and P-value of 0.000. Drape coeffi cient is the inverse of drapeability. As mentioned earlier, drape is the extent to which a fabric will distort when it is permitted to droop under its own weight. It is correlated with a fabric's mechanical properties. Th e most signifi cant factors include bending, shear, formability, fabric weight and thickness. Drape depends on a fabric's parameters, including structure, yarn type and fi bre content, as well as fi nishing treatments [14]. Table 4 shows a multiple linear regression equation of the fabrics' properties. Th ese equations would be useful in predicting crease and wrinkle recovery, as well as the drape coeffi cient of the fabrics. Th e adjusted R 2 value is an indication of the correlation of yarn properties (factors) and fabric characteristics (responses). When the adjusted R 2 value increases or decreases to 1 or -1, this indicates a strong correlation between them. In this study, the correlation of stiff ness and yarn parameters was low (Adj. R 2 = 0.231 for fl exural rigidity and Adj. R 2 = 0.125 for bending modulus). However, crease recovery, wrinkle recovery and the drape coeffi cient showed a very good correlation with the studied yarn parameters, with values of Adj. R 2 = 0.0998, Adj. R 2 = 0.975 and Adj. R 2 = 1, respectively. Th e rigidity (high drape coeffi cient) of woven fabric F 1 with a coarser count was higher, while fabrics with 36.4 tex had a higher drape coeffi cient (rigidity) of 74.8%. Th is indicated that the rigidity of fabrics increased proportionally with an increase in the yarn count (tex). In the case of a coarser count, the number of fi bres is high in the yarn cross section and leads to a high drape coefficient, making them less comfortable during wear. Because of its high drape coeffi cient, the fabric demonstrated less drapeability and fl exibility. It was observed that fabric thickness and weight also aff ected the drapeability of woven fabrics. Fabrics F 2 and F 3 , with a low thickness and low weight, demonstrated a better drapeability of 66.1% and 72.7%, respectively. A similar concept was also reported in earlier research [14]. Süle reported that woven fabrics with thicker weft yarns and higher weft densities had a higher rigidity [10].

Wrinkle recovery
Th e wrinkle recovery of woven fabrics made from four types of yarns with diff erent properties demonstrated a signifi cant change at an F-value of 30.917 and P-value of 0.000 (see Table 5). As the results show, wrinkle recovery was signifi cantly aff ected by yarn count, twist, elongation and the tenacity of yarns, respectively. A high level of wrinkling was observed in fabric F 4 made from a low twist, coarser count, and low elongation and tenacity of yarn. Th is was because when a load was applied on the fabric, it became compressed and deformed. When the applied load was removed, the tendency to recover to the original position was very low because the structure of low twisted yarns tends to be distorted. Th e individual correlation of yarn parameters was analysed using linear regression. Th e results showed that yarn elongation and tenacity were highly correlated to wrinkle recovery at Adj. R 2 of 0.928 and Adj. R 2 of 0.924, respectively. Yarn twist also had good correlation at Adj. R 2 of 0.879, but was low for yarn count at Adj. R 2 of 0.596.  . Th e wrinkle recovery grade was thus reduced from fabric F 1 , F 2 , F 3 to F 4 , respectively. Th is indicated that fabrics F 3 and F 4 with less twist yarns demonstrated poor recovery due to loose yarns in the fabric, which are easily damaged when a load is applied and the tendency of recovery to the original position is lower. However, woven fabric F 1 with a higher yarn twist and fi ne count demonstrated very good wrinkle recovery. Th is is because of the higher number of turns per meter, which contributes to resistance to the applied load (undamaged) and easier recovery. Th is result is confi rmed by a previous similar report on the recoverability of knitted fabrics [15] and the bending properties of woven fabrics [13].

Crease recovery
As evident in Table 5, the four types of woven fabrics made from yarns with diff erent properties demonstrated no signifi cant change in terms of the statistical value of the F-value of 3.546 and P-value of 0.069. A P-value is ≥ 0.05 indicates that there is no diff erence between samples. As seen in Figure 3, the crease recovery of the F 3 and F 4 samples seems to have greater value. However, that value was statistically insignifi cant. To predict the crease recovery of fabrics, a multiple regression equation was developed and is presented in Table 4. All factors together had an Adj. R 2 of 0.998. In addition, linear regression was also performed to show the relation of a single factor and response (crease recovery). Yarn twist (Adj. R 2 of 0.991) elongation (Adj. R 2 of 0.973), yarn count (Adj. R 2 of 0.955) and tenacity of yarn (Adj. R 2 of 0.854) were correlated with the crease recovery of fabrics. Th e Adj. R 2 value was used to analyse individual factors, while the remaining yarn parameters were taken as constants.

Conclusion
In this study, four types of woven fabrics were produced from yarns with diff erent properties by varying the weft yarn count, twist, strength and elongation. In all woven fabrics, the same warp yarn properties were applied, with the only variation in weft direction due to diff erent weft yarn properties. Th e eff ect of these yarn properties on the psychological comfort of woven fabrics was studied and analysed statistically. To predict the value of crease recovery, wrinkle and the drape coeffi cient of fabrics, an equation was developed using multiple regression. A statistical analysis showed that yarn twist, count, tenacity and elongation demonstrated insignifi cant changes at an F-value of 3.546 and Sig-value (P-value) of 0.069 on the crease recovery of fabrics. Th e stiff ness of fabrics, expressed as fl exural rigidity and bending modulus, also showed an insignificant diff erence between the samples at an F-value of 38487.969, Sig-value of 0.057 and an F-value of 25.506, Sig-value of 0.055, respectively. However, the wrinkle properties and drapeability of fabrics were signifi cantly aff ected by yarn twist, count, tenacity and elongation. Woven fabrics F 3 and F 4 with less twist yarns had poor recovery properties due to loose yarns in the fabric, which were easily damaged when loaded, while woven fabric F 1 with a higher yarn twist (937.4 m -1 ) and fi ne count (21.5 tex) had very good wrinkle recovery properties. Due to a coarser yarn count, a high drape coeffi cient (poor drapeability) was observed because the number of fi bres is high in the yarn cross section, thus making them less comfortable.