Indi an Journal of Fibre & Textile Research Vol. 29, March 2004, pp.69-73

Studies on abrasion resistance of silk fabrics

B V Vaslimathi . T I [ Somashckar", S L Chil akwad & G Shrikanth

Central Sil k Technolog ical Research Institute, BTM Layout, Madi vala, Bangalore 560068, lndia

Receil'ed 16 Jllly 2002; Revised received 9 December 2002; accepted 18 Decell/ber 2002

Around 100 different sampl es of commerc ially popul ar sil k fabrics were collected and tested for fabric constructional particulars and two functional properties, viz abrasion resistance and bursting strength . The silk fabr ics were tested for abrasion resistance on both Taber type and Martindale testers and the end point of the test was determined as per the pin­hole and wei ght loss methods. Abras ion resistance, expressed in terms of number of cycles to form a pin-hole, showed better consistency in terms of various inter- re lati onships. The interdependence of factors influencing the abras ion resistance wi th other functi onal properties have also been deduced.

Keywords: Abrasion resistance, Martindale tester, Silk fabric, Taber tester IPC Code: lnt. 0 7 D06H 3/00

1 Introduction One of the most important physical properties of a

textile fibre in relation to durability is its ability to withstand abrasion . Fibres with poor abrasion resistance break and splinter, leading to the formation of holes in the fabric) . Abrasion of fabrics may take place in normal usage when they are flat , folded or curved such that the fibres are subjected to three types of abrasion, namely fl at, flexed and edge. Prediction of performance in relation to abrasion is considered to be difficult as many factors influence abrasion) . Some fibres have better abrasion resistance than do others. Yarn construction is also an · influencing factor. Loosely twisted yarns abrade more easily than do tightly twisted yarns. In the loosely twi sted yarns, the individual fibres are more likely to be subjected to being pulled out from the body of the yarn onto the surface of fabric2

•

In woven fabric construction, the aspect of crimp distribution seems to be particularly important. The crown, a part of the yarn that protrudes above the surface of the fabric, is subjected to the pressure and surface or edge abrasion. If there are more crowns of the same height in a given area of cloth , the wear will be distributed over the surface of the fabric more evenly 3 . Fabrics made from uneven loosely twisted slubbed yarns with higher unevenly spaced crowns have poor abrasion resistance ).

aTo whom all the correspondence should be addressed . Phone: 26685238 ; Fax: +91-80- 26680435 ; E-mail: cstri @vsnl.net

Some abrasion effects are immediately recognizable. Where the rubbing is general over a large area, there may be a thinning of cloth and a loss of strength , leading eventually to bursting or tearing. If the wear is concentrated, the first effect to be noticed is likely to be a hole. If the cloth is square in construction, i.e. having equal numbers of similar yarns in both the warp and weft, it is possible that the hole will be formed by the destruction of both sets of threads in the same area. ]f,

however, one set of threads is predominant on the surface of the cloth, as in the case of dupion sil k fabrics, these threads are abraded preferentially and the hole will be seen to be crossed by a ladder of the opposing threads. Such a hole may be the starting point of a tear if the cloth is sufficiently stressed in that area:!.

The reports4-7 on the abrasion resistance of a wide

range of fabrics do not include silk. However, si lk is rated as one with poor abrasion resistance. It is found that the problems associated with the characterization of textiles in respect of abrasion resistance are serious and by no means solved I.2.

The imitative test for wear :-esistance is branded as a snare and a delusion8

. It is not required that service li fe be forecast in same unit of time, but a relati ve comparison of one fabric with another would be more meaningful9

.

In the present work, around 100 different samples of commercially popular silk fabrics were collected and two tested for fabric constructional particulars and two functional properties, viz abrasion resistance and bursting strength. The silk fabrics were tested for

70 INDI AN 1. FIBRE TEXT. RES ., MARC H 2004

abrasion resistance on both Taber type and Martindale testers

2 Materials and Methods Around 100 samples of various types of

commercially popu lar silk fabrics were collected and tested for fabric constructional particulars and two of the functional properties, viz abrasion resistance 10 and bursting strength. The range of fabric particulars of these samples are given in Table I . Crimp ratio (warp crimp/weft crimp) has been used instead of absolute va lues of warp crimp and weft crimp s ince the re lative maonitude is what is know n to influence the abrasion

"" properties of any fa bri cs. The two philosophies, one of duplicating actua l

condition s of wear and the other of selecting the most importan t causes of wear and corre lating them wi th service tests, have led to the development of more

. I ' II than 50 abrasion and wear tes ting mac ll11es . In the present study., the si I k fabrics were tested for

abrasion res istance on two types of abrasion testers, viz Taber type and Marti ndale tes ters. The end po int of the tes t was determined as per the pin -ho le method and weight loss method. In the pin-ho le method, the fabric was abraded till a hole was fo rmed on its surface, while in the weight loss method, the fabric was abraded for 50 cycles. The initial weight and the final wei oht were determined for calculating the per

"" cent weight loss.

2.1 Marli ndale Tesler

The tester is des igned to ensure that the speci men under tes t is subj ected to rubbing and fl ex ing from all directions. The pattern traced by the ab rad ing heads is the well-known ' Lissajous' figure. An area of 100 cm2 was subjected to abrasion in all the directions. It was used to compare the resistance to ab ras ion of different types of cloths.

2.2 Taber Type Tesler Material s to be tes ed are subjected to wear action

of two abrasive wheels at a known pressure. The

abras ive / wear action results when a pair of abrasive wheels is rotated in opposite direction s by a turn tab le on which the specimen material is mounted.

An excl usive feature of the Taber abrasers is a patented "X" pattern of abrasion, produced by a rotary rub-wear action of the wheels . The wear pattern formed is that o f two intersect ing areas o r a slightly curved herringbone configuration fr m the outside to the centre and from the centre to the outside of the specimen . An area of 30 cm2 was subj ected to test and a complete c ircl e on the material surface was abraded at all angles of grain or weave. There was provi s ion fo r vacuum pi ck-up for remov ing ab raded parti cles from the urface of the fabri c.

The test results were later analyzed statisticall y to study the inter-relationships and to draw inferences. The analyses were carri ed out variety wise and the poo led data across all the vari eti es of si I k fabrics have been referred to as 'Overall '. The following observations were made:

(i) Corre lat ion of the abras ion tes t results of Martindale and Taber type testers was studi ed.

(ii ) Correlati on between the abras ion tes t results was also determined on the basis of weight loss and pin-hole methods.

(iii) The abrasion test results were correlated with different fabric parameters, viz wei ght/m2

, thickness, cover factor, e tc .

( iv ) 't' tests were carried out to es tabli sh the s ion ificance of di ffe rences in the abrasion res istance

"" of the most common varieties of silk fabrics.

3 Results and Discussion The abrasion res istance values of various types of

silk fabrics tested are given in Table 2.

3.1 Infl uence of Method of Test Abrasion resistance, ex pressed in te rms of number

of cycles to form a pin ho le, shows better consi stency in terms of various inter-relationships, while the weioht loss method does not show any trend . The

"" probable reasons for the lack of correlation in the case

Table I- Range of fab ri c parameters for different s ilk fab ri cs

Fabric Endslinch Picks/i nch C rimp ratio Warp denie r Weft denier We ight, g/m2 Cover fac tor T hi ck nes , mill

Soft s ilk 100- 129 92 - 127 1.2 - 8.1 18.3 - 38.5 18.2 - 56.4 18.2 - 56.4 11.2 - 18.9 0. 12 - 0.15 Taffeta 120- 248 90 - 115 4.0 - 14.6 29.8 - 36.4 65. 1- 100.9 51.5- 77.9 17 .5 - 23.6 0. 13-0. 17 Dupi on 92 - 240 S9 - 95 3.5 - 16.5 23.5 - 44.6 155.2 -4 11.5 75.5-134.0 16.6 - 24.4 0.15 - 0.36 Chiffon 138 - 223 64 - 11 0 0.1 - 0.6 17.6 - 22.3 18.6 - 46. 1 20.0 - 34.8 12.0 - 17.2 0. 12 - 0.18 Georgette 102 - !1 2 82 - 95 0.5 - 1.6 18.9 - 43.8 20.5 - 37.6 17.0 - 36.2 10.7 - 14.6 0.12 - 0.16 Crepe 108 - 141 69 - 129 0.4 - 1.3 34.6 - 87.9 33.6 - 81.0 32.0 - 65.9 13.9 - 19.5 0.13 - 0.20 SateenfTwill 120 334 103 155 0.9 1.3 35.3 - 52.8 36.1 - 62.9 5 1.6 - 92.5 17.6 - 27.7 0.12 - 0.25

VASUMATHI el al.: ABRAS ION RES IST ANCE OF SILK FA BRICS 71

Table 2-Abrasion resistance of va rious sil k fabrics

Fabric Number o f cycles to Weight loss (%) after 50

from a Ein ho le c~cles

Martindale Taber Martindale Taber

Soft sil k 52.7" 38.3" 0.56 2.65 Dupion 29.8b 2 1.1 h 0.57 3.96 Taffeta 63.3c 56.8" 0.28 0.93 Crepe 27.6h 18.8b 0.96 1.60 Georgette 9.0d 7.0d 0 .2 1 2.83 Chi ffo n II.Od 10.2d 1.26 2.64 Twi II/Sateen 85.0e 77 .0e 0.60 1.60 Overall 50.6 33.9 0 .63 2.32

J.b.c.d & e indicate signi fica nt di fference at 5%

Table 3-Corre lation between Martindale cycles and Taber cycles

Fabric

Soft sil k Dupi on Taffeta Crepe Overall

Correlation coeffic ient ( r )

0.82" 0 .92" 0.92 " 0.47 b

0.79"

" Significant at 1%: b Signi fica nt at 5%.

of weight loss were identified to be: (i) the size of the sample prepared for the tes t is much larger than the size of the specimen actually abraded, resulting in the di fference in weights, especial ly in the case of Taber tester, (ii ) there is no sucti on dev ice in Martindale tester to suck the abraded fibre particles from the surface, thereby allowing these particles to clog onto the surface of either the abrader o r the abradant , and (i ii ) since silk has comparati vely low abras ion res istance, it continues to abrade the surface fo r a definite number of cycles (say 50) even after a hole is fo rmed and destructs the fabri c to a greater extent , sometimes di sintegrating the fabric structure. Thi s leads to the generation of a large quantity of lint which gets embedded onto the fabric surface. Therefore, it was decided to use the abrasion res istance results obtained from the pin-hole method for further analysis.

3.2 Influence of Type of Tester Table 2 shows that the abrasion res istance

measured on Taber tester in terms of cycles to form a pin hole is lower than the corresponding results obtained on Martindale tes ter in all the varieties. The reason could be that the two abraders act on the fabric surface simultaneously in case of both Taber and Martindale testers and the fibres separated from the fabric surface due to abras ion gets clogged onto its surface, thereby delay ing the end point.

There exi sts significant correlation between Martindale and Taber in the case of abras ion

res istance expressed in terms of number of cycles to form a pin hole (Table 3). It is also observed that the weight loss on the Taber tester is much more than that on the Martindale tester. One of the reasons fo r thi s is the presence of vacuum suction arrangement in Taber. The abraded f ibre particles are duly sucked away from the fabric, while thi s has to be done manually in the case of Martindale tester at regular intervals.

3.3 Influence of Variety of Fabric Chi ffon and georgette fabrics show the leas t

res istance to abras ion fo llowed by the crepe fabrics. Chi ffon and georgette have high twisted yarns in both warp and weft , while crepe has high twisted yarn on ly in the weft. It is observed that the high twi st sti ffens the yam to a point where very little contact is made between yarn and abras ive. This, in turn , results in hi gh locali sed abras ive pressure and early breakdown of yarn structure4

. It has been shown that an optimu m twist is recorded beyond whi ch the increase in number of twists results in lower abrasion res istance9

.

The plain fabrics exhibit lesser abras ion res istance than taffeta, while the abras ion res istance of dupion silk fabrics is lower than that of pl ain fabrics. The reason for the low abras ion res istance of dupion fabrics could be attributed to the greater imbalance in the construction of fabric ow ing to the huge difference in the deni er of dupion yarn used as weft and reeled sil k yarn used as warp. This also results in greater di fference between warp crimp and weft crimp. Increased fabric thickness and larger yarn di ameter are generally related and provide marked improvement in abrasion res istance of tex tile structures4

. However, the largest diameter of the abrasion bearing yarns must be accompanied by yarn uniformity if increased wear li fe is des ired. This is not so in the case of clupion sil k fabrics. It is reported that the non-uniform heavy yarns actually cause fabric degradation because of the high pressure concentration at the ir crowns.

The better abras ion res istance of taffeta can be attributed to the higher warp density. It has been shown that when all other factors are taken constant, the abras ion res istance of warp flu sh fabrics is improved by increasing the number of warp crowns per square inch of the fabric4

. It is also stated that the warp texture should no t be increased to the point of jamming the warp threads and reduc ing the fle xibility of fabric structure.

Twill and sateen silk fabrics show the highest abras ion resis tance. Thi s reiterates the fact that weave has an influence on the abras ion resistance of fabrics.

72 INDlAN J. FIBRE TEXT. RES., MARCH 2004

Table 4-Correlation between abras ion cyc les and fabric parameters

Parameter Soft Silk Dupion

Martindale Taber Martindale Taber

Weight. g/m2 0.49" 0.60 " 0.49" 0.51 " Thickness, mm 0.25c 0.29 c 0.09 c O.IO c

Crimp ratio 0.13c 0.14c -0.3 1" -0.24 c Cover fac tor 0.6 1" 0.62b 0.75 b 0.76b

Ends/inch 0.52c 0. 47 c 0.S3b 0.S2b Warp denier OASc 0.6 1h 0.09 c 0.13 C

Bursting strength, kg/c m2 0. 14c 0.4S" 0.34" OAI b

"Significant at 5%; bS ignificanl at 1 %; cNot Sign ificant

Thi s is in agreement with the earli er observations4. In

fabrics with floats, the yarns and fibres are more free to move and are less consistently exposed to abradants. For this reason, twill and sateen fabrics may show better abrasion resistance. However, the longer floats do have adverse effect on the abrasion resistance of fabrics as already observed in the case of some jacquard designs 1.3.

3.4 Influence of Fabric Parameters Con'elation studies between abrasion resistance and

fabric parameters were carried out for the four varieties of ilk fabrics, viz plain, taffeta, dupion and crepe , which showed moderate abrasion resistance. The varieties at the two extremes, viz. chi ffon/georgette with very low abrasion res istance and sateen/twill with very high abrasion res istance, were excluded . The abrasion res istance of these fo ur vari eties of fabrics shows significant dependence on weight, thickness, cover factor and bursting strength of the fabric (Table 4). However, there are some exceptions.

Wherever the thickness of the fabric is primarily due to the coarseness of either the warp or the weft resulting in an imbalance of the structure, a higher fabric thickness is not contributing to abrasion resistance as can be observed in the case of dupion fabrics . Here, the increase in thickness is largely due to the greater imbalance in the crimp of warp and weft because of the coarser weft used and a higher warp density. The abrasion res istance in this case is not showing any improvement with the increase in fabric thickness while in the case of taffeta, where the increase in thickness is due to a closer construction (i ncreased sett in the fabric), the abrasion resistance is positi ve ly correlated with thickness.

The abrasion resistance is related to the thickness or diameter of that part of the tex tile structure which is exposed at the rubbing surface. For plain fabrics with high warp crimp and low weft crimp, the rate of damage is higher. Non-u niform yarns actually cause

Taffeta Crepe Overall

Martindale Taber Martindale Taber Martindale Taber

O.77 b 0.S3b 0.3S c 0.S3b 0.32h 0.33b

0.65" 0.74b O.IS c 0.50 " 0.25 " 0.21 " 0.06 c 0.07 c 0.25 c O.77b -0. 19 c -0.1 3 c 0.S3 b 0.94b 0.37 c 0.S5 b 0.29" 0.35" 0.S5 b 0.94b 0.61 " 0.50 c 0.3 1h 0.20 c

0.09 c O. OJ C 0.69" 0.79b OASb 0.61 " O.77 b 0.74b 0.07 c 0.74b 0.26" 0.52h

fabric degradation because of the high pressure concentration at their crowns4

.

There is a significant correlation between cover factor, weight and abrasion resistance in all the cases. The dependence of fabric abrasion res istance on the cover of the fabric and its weight is established beyond doubt in all the sorts tested.

Incidentally, there has been a significant relationship between bursting strength and abrasion resistance in almost all the valieties tested (Table 4). Bursti ng strength is the composite strength of both warp ancl weft and it indicates their ability to withstand the multi­directional loading. Abrasion is recognized as a seri es of repeated application of stress and therefore a capacity to absorb the stress is required on the entire fabric structure including the fibres. T his similarity could be the contributing factor for the interdependence of these two factors across all the varieties of silk fabrics tested.

Multiple regress ion analysis was carried out to estimate the strength of relationship between the abrasion cycles and all the fabric parameters. The mu ltiple regression equations along with the R2 values are given in Table 5. The explanation for the significant relationships observed in the analysis can very well be explained in the context of the findings of the earlier studies4

. The variety-wise explanations are as given under:

Soft Silk Fabrics-The abrasion resistance of soft silk fabric is observed to be a funct ion of its cover factor. Soft silk fabric being almost a square fabric, it is quite understandable that its cover fac tor significantly explains the abrasion resistance with equal role of warp and weft.

Dupion Silk Fabrics-Here, endslinch, crimp ratio and cover factor explain the abrasion resistance of the fabric. Dupion fabrics are weft predominant. The warp denier is much fin er than the weft deni er and the warp crimp is also much higher. It is observed4 that the

VAS UMATH I et al.: ABRAS IO N RESISTANCE OF SILK FABRICS 73

Table 5 - Mult iple regression o f Jbrasion cycles with fabric parameters

Fabric Tester Constant Endslinch Warp denier Soft Sil k MartindJle -69 .2

Taber - 18.1 Dupion MartindJle -9:l.4 0.35 1

Taber -9.28 0.255 Taffew Martindale 45.99 0.54

TJber -32. 81 0.442 Crepe Martindale - 1.888 0.60'+

Tabt:r -45.424 02 11 Overall Martindale -228. 13 0.331

Taber -72.605 0.50'+

"Significant at 5%; "Significant at 1%

crimp distribution plays an importan t role in the abrasion res istance of fabrics and higher the cri mp imbalance, the lower wi ll be the abrasion resistance. The difference in warp and weft deniers will result in the warp yarns bending free ly while the weft yarns remain uncrim ped . Thi s cond ition wi ll lead to an actual reduction in geometric area of contact and to an earl y damage to the warp yarns. The damage can be delayed with more warp crowns in a g iven area of fabric . Weft, being a highly irregular yarn in spite of hav ing a larger diameter, is more prone to damage and this perhaps makes the cover factor as one of the paramete rs explaining the abrasion resistance a long with crimp ratio and warp densi ty.

Taffefa--Endslinch happens to be a sign ificant parameter explaining the abrasion resi stance. Taffeta is a warp flush fabric and the abrasion resistance of warp flush fabrics is known to be a function of warp density, other factors being the same~. It is observed that the higher number of warp crowns per square inch of the fabric reduces the normal load per crown, thereby imparting better abrasion res istance.

Crepe--Crepe fabrics have untwisted or low twi sted yarns in the warp and high twi sted yarns in the weft. It is reported 4 that the abrasion resistance is related to the diameter of that part of the textile structure which is exposed at the rubbing surface. In the case of crepe fabrics , the high twisted weft yarns re lax on finishing , imparting a crinkled and uneven surface to the fabric . Thus, the highly twisted yarns in the crepe fabric provide less contact with the abradant and the warp yarn takes the brunt of the abrasive action. It is confirmed 4 that the abrasio n resistance is related to the diameter of tex til e structure which is exposed at the rubbing sLllface and in thi s case it is warp.

4 Conclusions The abrasive wear of s ilk fab rics depends, to a

considerable extent , on the con :-~ ru c t i o n o f ya rn and

C ri mp ratio Cover factor Thickness R2

+ 7.2 0.3 1" + 3.3 0. 33"

-2.232 + 5.05 0.65 ~ 0.66 ~ 0.69 " 0.87 h

0.42" 3.178 0.69 ~

-5 .9 17 12 .67 0 .53 b

6.788 -22 1.0 0.6 1 "

fabric . Wear resi stance is more li ke ly to be a complex interaction of several properties whose re lati ve influence is different in different end uses .

The study has enabled ranking o f s ilk fabrics in the order of merit of the ir resistance to abras ion with chi ffo n and georgette having the lowest values and taffetta and twill s/sateens exhibiti ng the best abrasion resi stance. Abrasion of texti Ie fabr ics is a so co mplex mechani sm that the effect of individual structura l parameter is often masked by the interactions with each o ther. Thi s is we ll reflected in the difference in behaviour of diffe rent varieti es o f si lk fabrics. The abrasion resistance of sil k fab rics can be increased by increasing the geometrica l area of contact between the cloth and the abradant which can be done by usi ng the hi gher cover factor and eq ual cro wn heigh ts in warp and weft. However, the exact nature of influence of these parameters becomes specific to the type of fabric that includes a ll possible s tructural variations and the ir interact io ns.

References I Tortora Phyili s G, Ullderstalldillg Textiles (Mac mil lan

Publi shing Company , New York ). 1987. 48 1. 2 Tay lor H M, Textiles , 7 ( 10 78) 36. 3 Lord P R & Mohammed M H. Weal'il!g: COll versioll (!f Yam

to Fabric (Marrow Publishing Co Ltd. Eng land). 1982. 175. 4 Stanley Backe r & Tanenhaus S J. Text Res J, 2 1 ( 1951 ) 635 . 5 Stout E E & Moseman M B. in Textile Fibres. Yams (llId

Fabrics-A COlllparatil'e SlIrl'ev of tlieir Belwl'iollr \\'itli Special Rej erel/{'e to Wool , edited by E R Kaswe ll (Re inhold Publi shing Corporatio n, New Yo rk). 1953, 3 19.

6 Peirce F T , J Text IlI st , 28 ( 1937) 181. 7 Lo max J. J Text /11 .1'1, 28 ( 1937) 218 . 8 Booth J E, Prillciples '!f Textile Testillg (Butterworth

Sc ienti fi co London), 1983 , 266. 9 Hicks E M & Scorggie A G, Text Res J. [8 ( 1948) 4 16. 10 ASTM D3884-92 and IS 12673 : 1989 I I Kaswell E R, Textile Fibres. Ya m s alld Fabrics-A

COlllparative SlIn'ey (!f tli eir Beliw'iour witli Special Referellce 10 Wool (Re inhold Publishing Corporation, New York) , 1953, 30 1.