3. Results and Discussion
One of the reasons of the differences in fabric properties of woolen and regenerated wool can be the different morphological and physical structure of the fibers. The thickness of the woolen fibers was approximately 34.43 mm, while that of the regenerated woolen fibers was 43.53 mm, but dispersion of the regenerated woolen fiber thickness was higher than that of the natural wool. The length of the natural woolen fibers was 47.00 mm, and the average length of the regenerated woolen fibers was 46.25 mm; however, the length of the regenerated wool varied in the range of 24.00–91.00 mm, i.e., the length of the fibers was very uneven. The reason is that the fiber of regenerated wool is damaged in the converter machine, some of fibers were broken, whereas the other fibers remained intact. Because of that, the physical parameters such as fiber thickness and length become uneven as discussed in the references [
3,
4,
5].
The differences between the natural and regenerated woolen fibers can also be seen in the SEM images in
Figure 3. The flakes of the natural woolen fibers (
Figure 3a,b) are brighter, more pronounced, and more embossed than the flakes of the regenerated wool (
Figure 3c,d), which are duller and distributed closer to the fibers. This could be because the regenerated wool underwent significant mechanical and chemical effects during the converting process. As a result, the flakes would have rubbed off and stuck to the fibers, becoming less expressed than in the natural woolen fibers.
As mentioned above, fabric shrinkage in the directions of the warp and weft after finishing is very important because the final dimensions of a blanket depend on it. Therefore, it is very important to know the shrinkage in these directions after finishing before manufacturing the final product.
Seeking to compare the results of the fabrics from regenerated and natural wool, column diagrams of shrinkage in the directions of the warp and the weft were drawn (
Figure 4). The highest shrinkage in the direction of the weft occurred equally (–9.38%) for sample 2 (diamond twill weave) and sample 5 (derived basket weave) in Series 1 products with regenerated wool. Although the weave type of these two fabrics differed, floats of similar length (through three and four threads) predominated in the direction of the warp, leading to greater shrinkage in the direction of the weft. The lowest shrinkage in the direction of the weft in fabrics of Series 1 occurred for sample 9 with a derived twill weave (–3.75%). This weave had short floats in the directions of the warp and weft, i.e., floats through two and three threads predominated, and interlacing of one float in the weave existed. The highest shrinkage in the direction of the warp was obtained for sample 8 with a derived twill weave (–6.31%), which had especially long floats in the direction of the warp, while the lowest shrinkage occurred for sample 11 with a honeycomb weave (–2.91%), in which short floats and a high number of intersections in the directions of the warp and weft predominated. The authors investigated the shrinkage of fabrics woven in plain, different twill and basket weaves in the reference [
16]. It was established that the shrinkage of woven fabric in both weft and warp directions depends on the length of weave floats. Woolen fabrics without surface treatment undergo major changes in their structure including the shrinkage after washing; thus, the different treatments can be used to reduce these undesirable properties [
17,
18,
19,
20]. Thus, the main result of our investigation is confirmation that the shrinkage of woolen fabric depends on fabric structure, especially on the fabric weave and the length of floats in the weave.
It can be seen from the diagram (
Figure 4) that shrinkage in the direction of the weft was, in almost all cases, higher than in the direction of the warp—from 4% for weave 8 to 56% for weave 2. This could be because the weft threads were less tensioned than the warp threads and because the freer weft threads had the possibility of shrinking more than the tensioned ones. The ratio of shrinkage of weaves 1–6 and 10–12 was similar in the directions of the warp and the weft, i.e., approximately 50%. Shrinkage of weave 7 in the direction of the weft was also similar (–8.75%), but the ratio with warp shrinkage was lower than that of the earlier investigated weaves. This ratio of weave 8 was even lower—4%. However, the shrinkage tendencies of weave 9 in the directions of the warp and weft were opposite to those of other weaves, i.e., weft shrinkage was lower than warp shrinkage by 36%. These three weaves (7–9) were diagonals and their data differed from the data of the other weaves. This could be influenced by longer floats of one type in the direction of the warp and higher steps of the weave—the floats in the direction of the warp were longer than these in the direction of the weft, and because of this, shrinkage in the direction of the warp is easier. The authors of the references [
10,
16,
18,
19,
20,
22] investigated the shrinkage in weft and warp directions of different simple weaves such as plain, different twills, basket. They established that the weaves with longer floats in the directions of the warp and weft had the higher shrinkage than the fabrics with shorter floats. It was also established that in all cases, the shrinkage in the direction of the weft was higher than the shrinkage in the direction of the warp. These results are similar to those of our research, i.e., the shrinkage in both directions of the fabric depends on the length of the length of the weave float.
It can be seen from the diagram in
Figure 4 that the highest shrinkage for Series 2 in the direction of the weft was obtained for sample 1, woven in a diamond twill weave (–10%). Therefore, the highest shrinkage in the direction of the weft occurred for the diamond twill weaves in the samples of Series 1. The lowest shrinkage in the direction of the weft was obtained for sample 9, as well as for those fabrics with a diagonal weave (–5.94%). The highest shrinkage in the direction of the warp was obtained for sample 8 with a diagonal weave (–6.8%), as well as for the fabrics of Series 1. The lowest shrinkage in the direction of the warp in the fabrics of Series 2 was obtained for samples 11 and 12 with a honeycomb weave with the same shrinkage (–3.11%). This result also corresponded to predomination of the warp shrinkage of Series 1 in honeycomb weaves, which had short floats and a high amount of interweaving. The influence of the weave of woolen fabrics on their shrinkage in the directions of the warp and weft has been investigated by other scientists [
18,
19,
20]. It was estimated that the shrinkage increases when the float length of fabric weave decreases. The results of shrinkage can be compared to the results in [
16], where the new method of shrinkage establishment with a launderometer for plain, twill 3/1, and basket 2/2 weaves fabrics was presented, i.e., 47% higher values than those of the usual method. Only the preliminary shrinkage results for regenerated wool fabrics were found in the references [
1,
2,
3,
4,
5], but similar tendencies can be predicted for the regenerated wool fabrics as for the woolen fabrics.
When analyzing the shrinkage of Series 2 fabrics in the directions of the warp and weft (
Figure 5), the same tendency as in the case of the fabrics of Series 1 was highlighted, i.e., in all cases, shrinkage in the direction of the weft was higher than that in the direction of the warp—from 8% for weave 9 to 61% for weave 12. As in the case of Series 1, weaves 7–9 differed as well—their ratio of shrinkage in the directions of the warp and weft was the lowest, from 8% for weave 9 to 34% for weave 7. Warp shrinkage of these weaves was higher than that of the other fabrics because their long floats were longer in the direction of the warp than in the direction of the weft. Therefore, the magnitude of shrinkage in both directions became similar. The ratio of shrinkage in the directions of the warp and weft of weaves 10–12 was the highest and ranged from 44% for weave 10 to 61% for weave 12. These were fabrics of honeycomb weave, whose float lengths were the same in the directions of the warp and weft. Because of these reasons, weft shrinkage was higher than warp shrinkage because the weft was less tensioned than the warp during weaving. The properties of regenerated wool fabrics were investigated in the references [
2,
3,
4,
5]. The results do not contradict our results. The tendencies of shrinkage in the direction of the weft being higher than in the direction of the warp were established in references [
8,
14]. The reason is that the weft was less tensioned in the fabric and it had a greater possibility of shrinking in the fabric than warp threads after applying the finishing processes. Thus, one of the findings of this investigation is that the fabric shrinks more in the direction of the weft than in the direction of the warp.
Weaves with longer floats had higher shrinkage, while lower shrinkage was established for weaves with shorter floats and a higher amount of interlacing. This defined better stability of the fabric dimensions. The same results were established for different weave types of Series 1 and 2. The highest shrinkage in the direction of the weft predominated in the weaves of diamond twill, while the fabrics of derived twill with short floats and high interlacing were the most stable. The results in the direction of the warp were the same—the highest shrinkage was obtained for diagonal weaves, and the lowest for honeycomb weaves. Comparing the shrinkage of the fabrics of Series 1 to those of Series 2, it can be seen that the fabrics of Series 1 with regenerated wool in the weft shrunk less than these of Series 2 (100% woolen). Thus, the main finding is that to avoid less shrinkage, it is recommended to use more regenerated wool in the fabric. Such results do not contradict earlier results of the references [
10,
16,
18,
19,
20,
22] for the shrinkage in the directions of the weft and warp of woolen fabrics and the results of the properties of regenerated woolen fabrics [
1,
2,
3,
4,
5].
The finishing processes are also important for shrinkage in the directions of the warp and weft. Fulling and raising are the woolen fabric finishing processes that most influence fabric shrinkage. The results of fulling and raising shrinkage in the directions of the weft and warp for Series 1 fabrics with regenerated wool in the weft are presented in
Figure 6. Only one fabric was investigated from each weave group because the earlier results showed similar across weave groups. It can be seen from
Figure 6 that fulling shrinkage in the direction of the weft was the highest (from –16.00% for weave 3 to –16.56% for weaves 5 and 8) and the most expressed in comparison to shrinkage in the direction of the warp (from 0% for weave 3 to –3.91% for weave 5) and to raising shrinkage (from –0.74% for weave 5 to –4.38%) for weave 3 in the weft direction and from –1.13% for weave 3 to –1.76% for weave 8 in the warp direction). The results of the references [
10,
16] confirm that the woolen fabrics undergo the highest shrinkage during the fulling process because of wool fiber morphological structure, i.e., wool fiber is covered by scales, which caused fulling as well as shrinkage.
The fulling and raising shrinkage in the directions of the warp and weft of Series 2 100% woolen fabrics is presented in
Figure 7. It can be seen that the fulling shrinkage in the direction of the weft also differed and was the highest (from –10.00% for weave 5 to –13.5% for weave 3). The fulling shrinkage in the direction of the warp is also expressed for weave 5 (–6.94%). This is similar to the results in the references [
10,
16].
When comparing the fulling and raising shrinkage in the directions of the warp and weft for fabrics of different raw materials, it can be seen that the fulling shrinkage in the direction of the weft was higher for fabrics with regenerated wool in the weft, from 15% for weave 3 to 39% for weave 5. This can be explained by the fibers of regenerated wool being shorter and injured during the recycling process. Because of this reason, short fibers of regenerated wool fulled and shrunk more. The results of the reference [
16] are similar in that the fulling process had the highest influence on the shrinkage of woolen fabric. The authors of [
10] suggest the use of laser treatment for plain weave woolen fabrics to avoid high fulling shrinkage instead of usual methods of shrinkage reducing. To make these fabrics shrink-resistant, different types of finishing, such as oxidative, enzymatic using different chemical means (keratinase and papain [
17], protease [
7], dicyandiamide, glycolic acid, alkali, hydrogen peroxide [
8,
14,
15], Chlorine-Hercosett [
12,
13], laccase-assisted grafting of poly(tyrosine) [
18], POSS
® nanomaterial [
19], chitosan, wheat starch, and gum Arabic [
9,
20]), radiation, polymeric coatings, sol–gel coatings, plasma, and laser [
10] treatments, softening, water-repellent finishing [
21], and antimicrobial finishing [
21,
22], can be customized [
6].
In order to determine whether fabric weave and raw material (independent variables) influence fabric shrinkage (dependent variable), two-way analysis of variance was performed. The hypotheses are presented in
Table 2.
In order to check the hypotheses, ANOVA was performed. The results can be seen in
Table 3.
The results showed that the hypothesis “There is no difference in the average yield for any group of weaves“ was not confirmed because pvalue = 0.012 < 0.05 (significance level); however, the alternate hypothesis “There is a difference in the average yield of the group of weaves“ was confirmed. This means that the fabric weave influenced the shrinkage in the direction of the weft. The hypothesis “There is no difference in the average yield of the raw material of the fabric” was accepted because pvalue = 0.1578 > 0.05. Thus, the raw material did not influence the shrinkage in the direction of the weft. The hypothesis “The effect of one independent variable on the average yield does not depend on the effect of the other independent variable” was confirmed because pvalue = 0.8352 > 0.05. In conclusion, the shrinkage in the weft direction did not depend on the interaction between the weave group and the raw material of the fabric.
A multiple comparison test was performed in order to check the shrinkage in the direction of the weft between the four group of weaves. The
p-values for the hypothesis test showed that the corresponding mean differences equal to zero were
p1–2 = 0.8961,
p1–3 = 0.0103,
p1–4 = 0.3872,
p2–3 = 0.0414,
p2–4 = 0.7851, and
p3–4 = 0.2214; the
p1–2,
p1–4,
p2–4,
p3–4 values were higher than 0.05 (significance level) and it can thus be stated that the averages of these groups were the same statistically, but the
p1–3,
p2–3 values were less than 0.05; thus, the average of the third group was significantly different. It means that the weave influenced the shrinkage in the direction of the weft.
Figure 8 shows the multiple comparisons of the means.
A multiple comparison test was performed to determine whether the shrinkage in the weft direction differed between the two raw materials of the fabric. The
p-value =
0.1578 > 0.05; thus, the averages of these groups were statistically the same. This means that the raw material had no influence on the shrinkage in the weft direction (
Figure 9).
The same hypotheses were proposed during the analysis of the shrinkage in the direction of the warp. The results of the ANOVA can be seen in
Table 4.
The results showed that the hypothesis “There is no difference in the average yield for any group of weaves” was not accepted because p-value = 0 < 0.05. However, the alternate hypothesis “There is a difference in the average yield of group of weaves” was accepted. This means that the fabric weave influenced the shrinkage in the warp direction. The hypothesis “There is no difference in the average yield of the raw material of the fabric” was confirmed because p-value = 0.4906 > 0.05. Thus, the raw material did not influence the shrinkage in the direction of the warp. The hypothesis “The effect of one independent variable on the average yield does not depend on the effect of the other independent variable” was confirmed because p-value = 0.9947 > 0.05. Therefore, the shrinkage in the warp direction did not depend on the interaction between the weave group and raw material of the fabric.
A multiple comparison test in order to determine the shrinkage in the direction of the warp between the four group of weaves was performed. The
p-value for the hypothesis test showed that the corresponding mean difference was equal to zero;
p1–2 = 0.7858,
p1–3 = 0.0009,
p1–4 = 0.0551,
p2–3 = 0.0001,
p2–4 = 0.2776, and
p3–4 = 0.0000, and the
p1–2,
p1–4,
p2–4 values were higher than 0.05 (significance level), which means that the averages of these weaves groups were the same statistically. However, the
p1–3,
p2–3,
p3–4 values were lower than 0.05; therefore, the average of the third group was significantly different. This means that the weave influenced the shrinkage in the direction of the warp.
Figure 10 shows the multiple comparisons of the means.
A multiple comparison test was performed in order to determine whether the shrinkage in the direction of the warp yield differed between two raw materials of the fabric. The
p-value
=
0.4906 > 0.05; thus, the averages of these groups did not differ statistically. This means that the raw material did not influence the shrinkage in the direction of the warp (
Figure 11).
Thus, the ANOVA showed that the weave influenced the shrinkage in the directions of the weft and warp, but the raw material had no influence on the shrinkage. These results agree with the earlier results of our investigation that the shrinkage depended on the length of the floats in the fabric weave. The ANOVA also confirmed that the shrinkage did not depend on the raw material because the same tendencies of the shrinkage were obtained for both natural and regenerated wool fabrics. These results are the new findings of this research.
As mentioned above, the air permeability after finishing is very important because the thermal properties of the blanket depend on it. Therefore, it is very important to know the air permeability after finishing, before manufacturing the final product.
To compare the results of regenerated and natural wool, a column diagram of air permeability according to weave type was drawn (
Figure 12).
It can be seen from the presented results that the fabrics of Series 1 had higher air permeability—from 4% for weave 1 to 13% for weave 3. For regenerated wool, the wool scale may be destroyed during the preparation. As a result, entanglement should be more difficult for the adjacent woolen fibers or yarns; this may be one of the reasons for the better permeability. Therefore, the latter had higher air permeability than the woolen fabrics. Woolen yarns made from fibers of non-equal lengths had greater inequality and different cross-sections, and this resulted in higher air permeability, as can be seen in the case of the regenerated yarns. Based on a comparison of the fabrics according to weave type, sample 5 with a derived basket weave and sample 8 with a derived twill weave had the highest air permeability (2227.78 dm
3/(m
2 s) and 2184.36 dm
3/(m
2 s), respectively). This air permeability was lower by 2% than that of sample 5, but was higher than of the other samples. Sample 4 of a derived basket weave and sample 3 woven with a diamond twill weave had the lowest air permeability (1857.04 dm
3/(m
2 s) and 1920.50 dm
3/(m
2 s), respectively). The highest air permeability was established for samples 7 and 8 with derived twill weaves in Series 1 (2241.14 dm
3/(m
2 s) and 2277.88 dm
3/(m
2 s), respectively), while the lowest air permeability was found for sample 4 with a derived basket weave (1983.96 dm
3/(m
2 s)) and for sample 10 with a honeycomb weave (2137.60 dm
3/(m
2 s)). Thus, it can be concluded that the combined twill weave had the highest air permeability and the derived basket weave the lowest. These results correspond to the results in references [
25,
26], where the influence of different fabric structure parameters on fabric shrinkage in the directions of the warp and weft was investigated. The air permeability for plain and herringbone woolen fabrics was higher than that for other different twill weaves as presented in the reference [
23].
The highest air permeability for the samples in Series 1 was obtained for sample 5 with a derived basket weave (2227.78 dm
3/(m
2 s)) because it is the weave in which long floats in the direction of the warp (through four threads) predominated; therefore, the air permeability was high. The air permeability can also be influenced by the warp floats being distributed in large elements through which air mainly passes. The air permeability was 2184.36 dm
3/(m
2s) in sample 8, woven with a combined twill weave, in which the number of floats in the direction of the warp predominated even for floats through six and four threads and in the direction of the weft through three and two threads; thus, the weave had long floats, especially in the direction of the warp, and this increased the air permeability. This weave had quite large weave elements with long floats, through which air could flow the easiest in the direction of the warp, as well as in the derived basket weave. The lowest air permeability was obtained for sample 4 of Series 1 with a derived basket weave (1857.04 dm
3/(m
2s)). However, this weave had smaller floats than sample 5. The weave also has a repeated repeat; thus, it is balanced very well and interlaces with one float in the warp that is often repeated. These aspects could have influenced the low air permeability. Low air permeability (1920.50 dm
3/(m
2s)) was also obtained for sample 3 woven with diamond twill. The given results of air permeability correspond to the studies published in references [
23,
24,
25], where it was established that the value of air permeability changes depending on the increase in interlacing and the length of the floats. This was the reason for longer floats creating larger pores (spaces) in the weave and higher air permeability. One of the main findings of our investigation is that the air permeability of woolen fabrics depends on the length of the floats in the fabric weave.
The highest air permeability was obtained for samples 8 and 7 with natural woolen raw material from New Zealand of Series 2. Sample 8, woven with a combined twill weave, had an air permeability of 2277.88 dm
3/(m
2s), i.e., it was higher by 4% than that in Series 1. Sample 7, woven with a combined twill weave, had high air permeability (2241.14 dm
3/(m
2s)) due to its similar weave properties to sample 8. The weave also had floats through six threads in many places, and elements in the direction of the warp were distributed in large areas. This increased the air permeability. The lowest air permeability of Series 2 was obtained for samples 4 and 5 samples. Sample 4, similar to Series 1, had low air permeability (1983.96 dm
3/(m
2s)), while an air permeability of 2127.58 dm
3/(m
2s)) for sample 5 with a derived basket weave was obtained. The air permeability of woolen woven fabrics depends on their structure parameters, including the weave [
24,
25]. The given results correspond to previously described results, which state that weaves with longer floats have higher air permeability. Results in the reference [
27] proved that the weave, weft yarn density, and finishing process influence the air permeability of woven fabrics. The 2/2 twill woven fabric, whose porosity was the lowest, had the lowest air permeability in comparison to other six combined weaves fabrics. These results confirm the dependence of fabric air permeability on the fabric weave.
The given results show that the results of air permeability have to be compared to the shrinkage in the direction of the warp (but not in the weft), i.e., the air permeability is higher when the shrinkage in the direction of the warp is lower. Thus, one of the main findings of our research is that the shrinkage in the direction of the warp had a greater influence than that in the weft direction when the air permeability is analyzed. In the reference [
26], it was established that the raw material, linear density of the warp and weft yarn, number of yarn twists, warp and weft density, warp and weft crimp, as well as the weave influenced the porosity and air permeability of woolen woven fabrics of plain and twill weaves.
A comparison of fabrics woven from woolen yarn (Series 2) and regenerated wool yarn (Series 1) indicated that the latter had higher air permeability. For regenerated wool, the wool scale may be destroyed during the preparation. As a result, entanglement should be more difficult for the adjacent woolen fibers or yarns; this may be one of the reasons for the better permeability, and therefore the fabrics with regenerated wool had higher air permeability than the woolen fabrics. With the use of blends with other fibers [
25,
26,
28,
29] and surface treatment [
30,
31,
33,
34], the air permeability can be controlled. It was established that combined twill had the highest air permeability and derived basket weave the lowest. The air permeability was higher for those weaves in which the long floats in the direction of the warp predominated. The fact that the warp floats were distributed in large elements, through which the air mostly flows, could also have influenced the air permeability.
The same hypotheses were proposed when the air permeability was analyzed. ANOVA was performed (
Table 5).
The results showed that the hypothesis “There is no difference in the average yield for any group of weaves” was accepted because p-value = 0.3942 > 0.05; thus, it can be stated that the weave did not influence the air permeability. The hypothesis “There is no difference in the average yield of the raw material of the fabric” was not confirmed because p-value = 0.0043 < 0.05. However, the alternate hypothesis “There is a difference in the average yield of the raw material of the fabric” was confirmed. Thus, the raw material influenced the air permeability. The hypothesis “The effect of one independent variable on the average yield does not depend on the effect of the other independent variable” was also accepted because p-value = 0.7003 > 0.05. Thus, it can be concluded that the air permeability did not depend on the interaction between the weave group and raw material of the fabric.
A multiple comparison test was performed to establish whether the air permeability yield differed between the four groups of weaves. The
p-value for the hypothesis test showed that the corresponding mean difference was equal to zero;
p1–2 = 0.8716,
p1–3 = 0.8628,
p1–4 = 0.8843,
p2–3 = 0.4396,
p2–4 = 1.0000, and
p3–4 = 0.4562. The
p-values were higher than 0.05 (significance level); thus, there was no difference statistically, which indicates that the weaves of the fabric did not influence the air permeability.
Figure 13 shows the multiple comparisons of the means.
A multiple comparison test was performed in order to see if the air permeability yield differed between the two raw materials of the fabric. The
p-value
=
0.0043 < 0.05; thus, the means were significantly different, which indicates that the air permeability yield differed across the two raw materials of the fabric (
Figure 14).
The ANOVA showed that the weave did not influence the air permeability, but the raw material did.
The shrinkage in the directions of the warp and weft and the air permeability did not depend on the interrelationship of the weave group and the raw material of the fabric. Thus, it can be stated that searching for relationships between shrinkage and air permeability is not appropriate.