Comparative Case Study on Adhesion of Three Common Sizing Agents to Cotton and Polyester Yarns

LI Ying, HU Xiao, CHENG Fei, ZHOU Mi＊, ZHU Puxin＊

(1. College of Biomass Science and Engineering, Sichuan University, Chengdu 610065, China; 2. School of Textile Engineering, Chengdu Textile College, Chengdu 611731, China; 3. Sichuan Zhukang New Material Co., Chengdu 611534, China)

Abstract: The adhesion between warp sizing and fiber was systematically studied by using the roving method. Cotton roving and polyester roving were sized with various concentrations between 0.01% and 2.50%of acid-thinned starch, polyvinyl alcohol and water-soluble polyester sizing, respectively, and tensile property of the sized roving was tested accordingly. The break force of the sized roving was considered as the adhesion force of the sizing to the fiber in the roving method. The experimental results show that the effects of sizing film strength and fiber strength on the adhesive force can be weakened when the concentration 0.5% of size paste is used, instead of 1% in the roving method. The size morphology in the sized roving and on the surface was observed through scanning electron microscopy, in the form of penetration and coverage of the sizing in or on the roving. On the other hand, the Young-Dupré equation was used to calculate the adhesion work. The advantages and disadvantages of roving method and the adhesion work method were compared. The adhesion obtained by both two methods reflects the rule of chemical similarity between warp sizing and fiber.

Key words: sizing agent; adhesion analysis; roving method; adhesion work

1 Introduction

Warp sizing can provide cohesion force between fibers in a warp yarn and form a film on the yarn surface, which can improve the strength and wear resistance of the warp yarn to meet the weavability requirements of the warp yarn, and to reduce the broken ends of the warp yarn during the weaving process, so as to improve weaving efficiency and reduce fabric defects[1]. Generally, the sizing agent forms two kinds of films after sizing: the thick one on the warp surface to resist friction and abrasion during weaving, and the other thinner ones flexibly binding the fibers together in the warp yarn, as schematically shown in Fig.1[2,3].The adhesive power of a sizing agent determines the sizing rate required for a warp yarn, and affects the physical and mechanical properties of the sized yarn.Correspondingly, it can exert a significant influence on the quality of the grey cloth[4]and the weaving efficiency of loom during the weaving process where the sized yarn is participated. Therefore, the adhesive power of sizing agents has received extensive attention[5,6].

Due to the requirement of film forming,textile sizing generally belong to organic polymers.Physicochemical interactions between the size agent/fiber pairs play important roles for their adhesive power[7]. If there is a certain chemical similarity between a sizing agent and the yarn to be sized[8], their adhesiveness can be achieved via interactions such as van der Waals force, and/or hydrogen bonding.Thus, the higher the molecular weight of the sizing agent the better its adhesion to the fiber. Since the intermolecular forces could accumulate with increase of the molecular weight on the one hand, and the size film with higher molecular weight has greater strength on the other hand[9]. For example, a small amount of high molecular weight polyacrylamide can greatly increase the film strength of a starch sizing agent, as a result of increasing the adhesion of the starch to fibers[10,11]. A sizing film should be tenacious and, more importantly, flexible or extensible, as it must follow the yarn repeated extensions on the loom and protect the yarn against multiple mechanical stresses[12]. Other factors such as surface activity of the sizing paste also exert significant influences on the adhesiveness. It is necessary for the paste to wet[13]and penetrate[14]the yarn during sizing, which involves adaptive surface energy of the fiber and the paste, and their polar and nonpolar interactions[7]. Moreover, superficial structures of the yarn[15], the diameter[16]and surface roughness of fiber[17]can be other physical factors, so the surface of the fiber or yarn may be modified by plasma[18]and photonic[19]radiations.

Adhesion measurement is critical to evaluate sizing materials. Several methods have been developed to evaluate adhesiveness between sizing agent/fiber pairs, including single fiber pull-out method[20], the evaluation of the bond failure between FRP (fiber reinforced plastic) composites and substrates[21], friction test[11,12], roving method[1]and interfacial chemistry method[22]. In the roving method, the adhesive force is expressed intuitionisticly and easily by the break force of the roving sized with 1% add-on of sizes[23,24].Before sizing, the roving with little twists of the yarn has the break force low enough to be negligible[11].When the roving is soaked into 1% concentration of a sizing paste and then dried, the strength of the roving(fiber cohesion) gives a sharp increase by tens of times,according to the sizing agents selected[25]. The roving method is not limited by the type and viscosity of sizing agents, but the wetting uniformity of the sized roving which has definite influence on the test results.It also has been noticed that the value of the adhesive force, as a comprehensive index of the test, is consist of the strength of fiber and roving[13], the cohesive force[15]of sizing agent and the adhesion of sizes to fiber. Although the roving strength is low, the cohesion of the size itself accounts for a large proportion of the breaking strength, and it is difficult to separate this part of proportion from the test data. Unfortunately, these influencing factors are so little known, thus they have aroused deep research interest.

In this study, the adhesive force of three kinds of commonly used textile sizing of starch, polyvinyl alcohol and water-soluble polyester size, was measured by using roving method. The influences of the tenacity of sizing films and the fiber strength on the adhesive data were systematically analyzed. Comparative analyses of the adhesion work were conducted by measuring surface tension of the sizing pastes and water contact angle on the sizing films. It is expected to have a meaningful understanding of the adhesion determination for warp sizing through this study.

2 Experimental

2.1 Materials

Acid-thinned starch (ATS, apparent viscosity 13.1 mPa·s at 5wt%) was supplied by Sichuan Zhukang New Material Co. (Chengdu, China); polyvinyl alcohol (PVA, with degree of polymerization 1700 and hydrolysis degree 99%) from Sinopec Chongqing SVW Chemical Co. (Chongqing, China); Water soluble polyester size (WSP) from Sichuan Hanzhou Chemical Technology Co. (Meishan, China); Analytical grade glycerol was from Kelong Chemical Reagent Co., Ltd.(Chengdu, China); Cotton roving and polyester roving about 1000 tex were provided by Sichuan Jinhua Co.;And deionized water was used in all of the experiments.

2.2 Adhesiveness test with the roving impregnation

The sizing content in the impregnation bath solution was fixed at 1% mass fraction with 2 500 mL amount of the bath solution for the test, and then heated at 95 ℃ for 30 min. Cotton roving or polyester roving was wrapped around a special metal frame.The wrapped roving on the frame was immersed in the cooked paste bath at 95 ℃ and hold for 5 min[26,27].After that, the sized roving was hanged to be dried at room temperature and balanced for 48 h in a controlled environment over the saturated solution of CuCl2(RH=68%) before evaluation. The break force(BF), elongation at break (EB) and break work (BW)of the roving before and after sizing was measured using a YG061 Tensile Tester (Laizhou, China) with a clamping distance of 100 mm and stretching speed of 100 mm/min[28]. For each data point, 20 samples were tested and the average value of BF was calculated as the adhesive force between the sizing and the fiber[28].

According to the above method, the tensile property of the sizing agents, ATS, PVA, and WSP,to cotton roving or polyester roving was determined,respectively. The test was conducted at different sizing contents of the impregnation bath solution, such as 0.01%, 0.025%, 0.05%, 0.1%, 0.25%, 0.5%, 1%, 1.5%,2%, and 2.5%, respectively.

2.3 Preparation and strength test of size films

Sizing pastes of ATS, PVA, and WSP with 6%solid content were cooked at 95 ℃ for 1 h in a threenecked flask, respectively. Each of the cooked paste of 60 mL was spread out with a syringe into a 210 mm×150 mm×3 mm glass trough in a multi-point manner, rested on a leveled surface and then was dried naturally[29,30]. The cast film about 0.1 mm thickness was carefully peeled away from the trough.

The film was cut into ribbons of 200 mm × 10 mm, the average thickness was measured by a CH-10-AT thickness gauge (Shanghai Liuling Instrument Co.,Shanghai, China) and equilibrated in a desiccator for 48 hours under the condition of RH 68%. The tensile properties such as tensile strength (TS), elongation at break (EB) and break work (BW) of films were measured by YG061 Tensile Tester. The holding distance of the sample is 100 mm and the stretching velocity is 100 mm/min. Ten pieces of the films were tested for each sample, and the average value of the results was obtained.

2.4 SEM observation of sized roving sections

The sized roving was clamped in a Y172 fiber auto slicer (Wenzhou, China) and was sliced along its cross section. The surface coating and permeation of the sizing were observed by Scanning Electron Microscope (Inspect F; FEI Company, Hillsboro,Oregon, USA).

2.5 Determination of surface tension of sizing pastes

A sizing paste with solid content of 2% was prepared by cooking at 95 ℃ for 30 min, and then cooled to 25.0±0.1 ℃ for determination. For ATS paste, 72.0±0.1 ℃ was selected and maintained by a circulating water bath to avoid retrogradation of starch solution. Surface tension was measured immediately by using a Sigma 703 Surface Tension Meter (Wilhelmy hanging plate method, KSV Company of Finland).

2.6 Measurement of contact angle

The cooked sizing solution of 2.0% was accurately prepared. The solution droplet volume of 4 μL was dripped onto a hydrophobic (polyester) film and a hydrophilic (cellulose) film, respectively, and the contact angles were measured by an OCA 20/6 system(Dataphysics Instruments GmbH, Germany) at ambient temperature. The average value of each sample was measured for 10 times.

3 Results and discussion

3.1 Comparison of adhesion properties of three sizing to cotton and polyester rovings

The tensile properties of sized cotton and polyester rovings with acid-thinned starch (ATS),polyvinyl alcohol (PVA) and water-soluble polyester size (WSP), respectively, are given in Table 1.

For cotton roving, it can be seen from Table 1 that BF order of the three sizing agents from high to low is PVA ＞ ATS ≥ WSP. Because BF value of a sized roving is generally used to express adhesiveness of the sizing material approximately[25], the BF order of the three sizes for cotton roving also reflects the order of their adhesion. The order of BW is PVA ＞ WSP ≥ATS, which shows the same order as EB.

For sized polyester roving, three values have the same order for the three sizing: WSP ＞ PVA ＞ ATS. In this sense, BW value of a sized roving is also one of the effective indicators for adhesion of the size agent.Compared with BF, BW value contains an elongation effect that can better reflect the flexibility factor contained in the real adhesion force[2].

The results in Table 1 show that the polyester roving sized with WSP has the highest BF value in all sized roving samples. This is mainly due to the similarity of chemical structures between polyesterfibers and WSP size, that is, both of them contain the same chain unit-ethylene terephthalate. Similar structures are compatible with each other[31]. On the other hand, it also shows that the BF values of the sized polyester rovings are generally greater than those of the sized cotton rovings, correspondingly. This indicates the influence of fiber and roving strength on the BF value. As the roving is sized, the fibers are connected by the size polymer and will not readily slip each other under stress. In that way the fiber assembly becomes a stress-resistant skeleton, and the strength of the sized roving is greatly enhanced[25]. Thus, it is not surprising for ATS in Table 1 that the BF of sized polyester roving(55.23 N) is higher than that of sized cotton roving(45.69 N). Although this does not conform to the law of chemical similarity of ATS with cotton, the fiber and roving strength must have played an important role in BF values of sized rovings, because the BF of unsized polyester roving (2.38 N) is higher than that of cotton roving (1.89 N). Therefore, the BF (adhesion) value of the sized roving must include the contribution of fiber and roving strength.

Table 1 Adhesion of three sizes to cotton and polyester rovings

3.2 Comparison of mechanical properties of the three size films

From the discussion above, it can be concluded that the test data of the adhesion of a sizing material by roving method is a comprehensive indicator of the roving strength, the adhesion force of the sizing material to the fiber, the cohesion of the sizing itself,and the fiber strength. Among all factors, the cohesion of the size will account for a large proportion of TS for the sized roving. It is well known that the cohesive force of a polymer material comes from intermolecular or interatomic force in this material. For this reason, the cohesive force of the size film is indirectly reflected by the tensile property of the film in the experiment. The results are shown in Table 2.

Table 2 Mechanical properties of size films

It can be seen from Table 2 that the order of TS from high to low is: PVA ＞ ATS ＞ WSP, which is the cohesive order of the sizes tested. The cohesion of a polymer is related to its molecular weight and molecular interactions. EB values in Table 2 can reflect the flexibility of size polymers. Among the three typical warp sizes, chemical sizes, such as PVA and WSP, have higher EB values in comparison with the very low EB of the ATS film. The break work (BW), the area under the stress-strain curve, can be used to characterize the strength and toughness of the polymer. PVA sizing has the highest TS and higher EB, so it has the highest BW value that is about 4 times of WSP and 130 times of ATS, as shown in Table 2. Therefore, we can say that PVA is a strong and tough sizing material, and on the contrary, ATS is a brittle one. It is noteworthy that the WSP size film, a flexible material, has the lowest TS but the highest EB value, and its BW value is only slightly lower than PVA and much higher than the ATS sizing material.

3.3 Analysis of the adhesiveness of the three sizing materials

In the data of adhesiveness of a sizing material to a fiber (BF in Table 1,i e, the break force value of roving sized with 1% sizing concentration)[26,27], the contribution of the fiber strength to the adhesiveness of the sizing material should be included in addition to the commonly considered factors such as roving strength,size film strength and the real adhesion between sizing material and fiber. Because of very low cohesive force between the fibers in the unsized roving, slippage occurs readily between the fibers under tensile stress,causing a very low BF value for the unsized roving.After sizing, however, the fibers are bonded by the sizing material, and have a great contribution to the roving strength as a supporting framework. In order to analyze the influence of these factors on the adhesion,we define A as the adhesion force of the size material without influence of roving strength,i e, the difference in break forces (N) of the roving before and after sizing with 1% sizing concentration (Table 1); B as the strength contribution of the unit cross sectional area of the size film to the adhesion,i e, the ratio (unit in mm2)of A value to the TS (Table 2); C as the contribution of the fiber strength to adhesion,i e, the ratio (tex)of A value to the fiber strength. According to Table 1 and 2, and fiber strength (the average value of cotton fiber is 0.31 N/tex; and the average value of polyester staple fiber is 0.55 N/tex), for the three sizing agents discussed in this paper, the calculated indexes of these factors are shown in Table 3.

It is difficult to discuss the three indexes (A,B, and C) independently for their strong correlations between them, but we are still trying to analyze these effects according to Table 3. For the adhesion factor A, we concerned about ignoring the influence of roving strength. In the sense of A, PVA had the highest adhesion to cotton; and WSP had the highest adhesion to polyester. The results are consistent with those of previous studies[2,32].

Table 3 Factor analysis of adhesion force in the roving method

As for the adhesion factor B, the influence of the strength of sizing films was emphasized. On the adhesion to cotton roving, the biggest influence happens in the sizing film strength of WSP, followed by ATS. It means that when WSP and ATS are used for sizing cotton yarn, the strength of the size film needs to be improved. Similarly, for the adhesion to polyester fiber, the analysis of factor B indicates that WSP is especially needed to improve their sizing strength.

The adhesion factor C takes into account the influence of fiber strength. For cotton and polyester rovings, the order of contribution of fiber strength to adhesiveness is basically the same as that of adhesiveness A, in which the adhesion of ATS to cotton is much higher than that to polyester roving.The adhesiveness of the two synthetic sizing agents to polyester roving is higher than that to cotton, and the C factor in adhesion of WSP sizing to polyester is the highest in all cases[33]. These data, particularly for the C factor, are more reasonable than those expressed by the simple roving strength[15].

The structural similarity between sizing and fiber is the essence of adhesiveness. Both of starch sizing and cellulose fiber have similar glucose units in their chemical structures[9], and polyester fiber and WSP sizing have similar structural unit of ethylene glycol terephthalate, so that they have good adhesion between the fiber and the sizing. From this point of view, the data in Table 3 show that the adhesion factor C which eliminates the effect of fiber strength is the most reasonable index.

3.4 Effect of size concentration on adhesion force

In the previous discussion of the study, it was found that with the roving method the adhesive force of a sizing to a fiber obtained by using 1% sizing concentration was affected by the roving strength, size film strength and fiber strength. All these factors are also affected by the sizing concentration of the roving method. Therefore, the adhesion test with the roving method under different sizing concentrations can be helpful to understand the adhesion test with the roving method. Fig.2 and Fig.3 show the effect of sizing concentration on the break force for the sized polyester and cotton rovings, respectively.

Fig.2 Relationship between sizing concentration and break force of polyester roving sized with ATS, PVA, or WSP

Fig.3 Relationship between sizing concentration and break force of cotton roving sized with ATS, PVA or WSP

The whole process of sizing includes wetting,spreading and film forming of a sizing on a fiber surface. After sizing and drying, the sizing retained between the fibers in the sized yarn can play the role of holding the fibers together, and the sizing film on the yarn surface can protect the yarn. The realization of these functions depends on the sizing concentration. It shows from Fig.2 and Fig.3 that the break force (BF)increases with the increase of the sizing concentration.When the sizing concentration is lower than 0.25%,there is negligible difference between the adhesiveness of each sizing to polyester and to cotton. In this region,the sizing film attached to the roving surface becomes too thin to cover the fibers as well as the roving surface. Therefore, both the influences of fiber strength and cohesive force of sizing films on adhesion are insignificant.

With the increase of sizing concentration, more sizing molecules can take part in covering the fiber surface and bonding fibers together, the thickness of sizing film coated on roving surface increases, and the contribution of sizing film strength to adhesion is also increased[34]. When the concentration reaches 0.5%,BFs of polyester roving sized with WSP (Fig.2) and cotton roving sized with ATS (Fig.3) have higher values than other sizings, respectively, which is consistent with the principle of similarity of chemical structure between the sizing and the fiber.

Behera’s group[1]and Sherilyn’s group[11]have claimed that among common sizing materials, PVA has the best performance in mechanical properties such as strength, abrasion resistance and flexural rigidity.In Fig.2, when the sizing concentration of PVA is increased from 0.25% to 0.5%, the BF of PVA for polyester roving has a sharp increase, far greater than that of cotton (Fig.3) in the same concentration. This could mainly result from better adhesion of PVA to polyester fibers than that to cotton. For cotton roving in Fig.3, the PVA concentration range corresponding to a rapid increase in BF is from 0.5% to 1.0%. On account of good film-forming and strength-toughness of PVA[11], the increase in the BF could be attributed to cohesion of PVA in its film rather than the adhesion of PVA to cotton. For this reason, the strength of the sized roving increases slowly after the sizing concentration of PVA, as well as ATS and WSP, is higher than 1%,as shown in Fig.2 and Fig.3, as a result of the fiber surface being essentially covered and most of the space between the fibers being filled with sizing material.Therefore, when the concentration of the sizing paste is higher than 1%, the strength of the sizing film and the fiber strength accounts for a large proportion of the measured adhesion value.

From the principle of chemical similarity, it is known that WSP sizing and polyester fiber have strong affinity each other because of their similar structure,but the strength of the sizing film itself is low (Table 2)so that the cohesion of the sizing film has little effect on the adhesion. As seen from the comparison between Fig.2 and Fig.3, the break force of polyester roving is always higher than cotton roving after sizing with the same sizes. That is to say, in the adhesive force of WSP sizing to polyester fiber measured by the roving method, high proportions of adhesion force and fiber strength exist, along with a small proportion of sizing film strength. For polyester roving sized with WSP in Fig.2, the break force is equal to that of PVA when the sizing concentration is 0.5%, but above this point, there was an obvious upward trend higher than that of PVA.WSP sizing has the characteristic of “high consistency and low viscosity”[35]. Even at high concentration,the fluidity and permeability of the sizing are good in addition to its good affinity to polyester fiber, so the adhesive force of WSP to polyester roving is always higher than the other two kinds of sizing materials.

Through the above analysis, it is found that the adhesion of sizing material to fiber is affected by sizing concentration. The sizing film strength and the fiber strength account for a part of the adhesion force data measured by the roving method, especially when the sizing concentration exceeds 0.5%. According to the conventional roving method[26,27,36], the break force of the roving sized with 1% sizing concentration is used to evaluate the adhesive property of sizing material. And then the influence of sizing cohesion and fiber strength cannot be ignored and should be realized.

3.5 Distribution of sizing materials on sized roving

Commonly used sizing is natural or synthetic polymeric compound[2,11,33,37], all the sizing pastes of which belong to non-Newtonian fluids. The concentration of the sizing paste is the major factor affecting the sizing add-on rate. An increase of concentration will lead to the increase in molecular density, which leads to the increase of intermolecular forces and viscosity. The rate at which viscosity increases with the increase of concentration is different for different sizing agents. The WSP sizing agent can be dispersed in water with coiled chains of molecules and a low viscosity[35], so it is suitable for the preparation of sizing paste with high concentration and low viscosity.

Fig.4 SEM images of polyester roving sized with WSP size of (a)and (b) 2.50%; (c) and (d) 0.25% size concentration

The dependence of adhesion force of a size on sizing concentration can be verified by the size distribution on the sized roving. In Fig.4, SEM images show polyester roving sized with 2.50% and 0.25%concentration of WSP sizing, respectively. In the image of Fig.4(a), the roving structure is compact and the darker part of the roving outer layer is coated with the sizing. In Fig.4(b) the sizing film between the fibers is clearly presented, which bond the fibers together. While in Fig.4(c) the image shows a loose and incompact roving. Although there is little sizing coating on the outer layer of roving, there is no sizing film formed between the fibers in the sized roving in Fig.4(d).

It is easy to understand that the roving sized with suitable sizing concentration has a compact yarn structure bonded by the sizing film, which is expected to resist friction during weaving[38]. On the contrary, for the sizing concentration as low as 0.25%, it is difficult to form film between fibers and cover the roving surface. Therefore, it is important to choose a proper sizing concentration to ensure good adhesion of sizing material to fiber, and to bind the fibers in the yarn.

3.6 Surface tension and contact angle of sizing pastes

Surface tension is an important property of a sizing. If the surface tension of the sizing paste is too high, it will be difficult for the paste to wet the fiber and permeate the yarn in the sizing, and the result is a surface sizing. In that case, the sizing film will be easily removed from the yarn surface, resulting in more hairiness on the yarn during weaving. If the surface tension of sizing liquid is too low, it will lead to excessive penetration of the liquid and insufficient protection of the warp surface. Although it may be easy to lease during weaving, it often causes more sizing shedding from the yarn bundles. A proper surface tension can ensure the balance between the sizing permeation and the surface coating, to hold the cohesive action between fibers, and to form the protective layer of the sizing film on the yarn surface.

Hydrophilic polymers, such as ATS, PVA, and WSP, have certain surface activity, as shown in Table 4. In the surface activity test, 2.0% sizing concentration was used, much higher than the critical micelle concentration CMC points (below 0.25%) of the three sizing pastes according to a pioneer test, so the stable surface tension values were obtained. Also, the wettability of size with 2.0% sizing concentration was studied on polyester film and cellulose film by contact angle measurement. Table 4 lists the contact angleθof the three types of sizing pastes to polyester film and cellulose film, respectively.

It can be seen from Table 4 that the order of surface tension of the three size pastes is: WSP ＞ PVA ＞ATS, and the contact angle of the three different sizes to cellulose film and polyester film is quite different.The lower surface tension of ATS paste than other sizes is due to the measurement conditions at a higher temperature. The contact angle of ATS to cellulose film is much smaller than that of polyester film. Therefore, it can be indirectly explained that the wettability of starch paste to cotton is much better than that of polyester.The wettability of WSP on polyester film is better than that on cellulose film.

Table 4 Surface tension, contact angle and adhesive work of size paste on fiber

3.7 Adhesion work of sizing paste to polyester film and cellulose film

Warp sizing is a process in which the sizing agent adheres to the yarn fiber, so we can use the adhesion work (Wa) to discuss the adhesion ability of the sizing paste to the fiber. TheWais defined as the work required for separation of the paste phase from the solid phase, or the negative value of the free energy corresponding to the adhesion at the solid/liquid interphase, which represents the interaction strength between the solid/liquid phases, given by Young-Dupré equation[22]:

where,Wais the adhesion work;γis the surface tension of size paste; andθis the contact angle of the paste on the solid surface.

By using theγandθvalues in Table 4 and according to Eq.(1), the adhesion work of the three types of size pastes to polyester and cellulose films can be calculated as in Table 4. From Eq.(1), it can be seen thatWais a thermodynamic variable that will not be affected by the size strength and fiber strength, nor by the molecular weight of the sizing. The data show that the adhesion work of PVA solution to cellulose film and polyester film is about the same; the adhesion work of WSP dispersion to polyester film is higher than that to cellulose film, and the adhesion work of ATS paste to cellulose film is much higher than that to polyester film.Therefore, in the sense of chemical similarity between size and fiber pairs, theWaresults have basic rules similar with the adhesion force from the roving method although the two methods have different mechanisms and quite different data sets.

4 Conclusions

Roving impregnation is a commonly method to evaluate the adhesive force of a sizing to a fiber.There are some influencing factors for the roving method that should be understood. In this paper, the influences of roving strength, fiber strength and sizing film strength on the adhesive force were analyzed and discussed. Three kinds of sizing agents, acid-thinned starch (ATS), polyvinyl alcohol (PVA), and watersoluble polyester (WSP), were used for studying the adhesion to cotton and polyester rovings, respectively.Based on these data and discussion, the concepts of the adhesion force of sized roving eliminating roving strength, the contribution of size strength per unit cross area of sizing film to adhesion and the contribution of fiber strength per 1000 tex to adhesion are put forward.By means of the roving method, the adhesiveness of sizing materials to fibers was measured under the condition of different concentration of sizing pastes. It was found that the adhesive force was affected by the sizing concentration, because of the sizing distribution variation on the sized roving surface and between fibers in the roving with the paste concentration.

Sizing concentration of 1% in the traditional roving method could be lowered, to decrease the effects of other factors such as the strength of the sizing film and the strength of the fiber on the adhesion. However,when the sizing concentration is as low as 0.25%, it is too small for the amount of the sizing to form an effective adhesive layer between fibers. As the size concentration is set to 0.5%, the influence of sizing film cohesion and fiber strength could be greatly reduced,giving a rather reasonable adhesion data.

Adhesion work of a sizing to a fiber can be optionally obtained either by the roving method or by surface chemistry method. The latter adhesion work is calculated by using Young-Dupré equation from the surface tension of the sizing paste and the contact angle of the size paste on the fiber surface. It was found that the adhesion work obtained by these two methods can supplement each other, and all of the two sets of adhesion work follow the adhesion principle of similarity in chemical structure.