Mechanical and thermal behaviors of ultra-high molecular weight polyethylene triaxial braids: the influence of structural parameters

Abstract

Recently, triaxial braids made from ultra-high molecular weight polyethylene (UHMWPE) have been recognized as one of the most popular composite reinforcements in the aerospace and defense fields. To further explore the mechanical characteristics of this material, a detailed experimental study on tensile behavior is reported in this paper. The triaxial braids show a “double-peak” phenomenon in tensile strength and deformation, caused by axial yarns and the in-plane shearing of bias yarns. The evolution of the braiding angle, measured during these tensile tests, is discussed according to the braiding parameters (initial braiding angle, number of axial yarns). Using the high conductivity properties of the UHMWPE material, the temperature caused by inter-yarn friction during tensile tests is also studied. This temperature is related to the evolution of the braiding angle. The temperature increases with the increasing number of axial yarns and decreases with increasing braiding angle. This study provides an experimental database on the influence of braiding parameters on the tensile behavior of triaxial braids.

Keywords

textile composites mechanical properties thermal analysis triaxial braids

Ultra-high molecular weight polyethylene (UHMWPE) is one of the most typical high-performance fibers, and it is widely used in the aerospace and defense fields, especially for ballistic-resistant body armor.^1,2 In addition, it also can be used in medical grade polymers in orthopedic applications, such as total-hip and knee-joint replacement components.³ Generally, fabrics can be obtained from UHMPWE mono-filament or multifilament fibers using weaving, knitting and braiding technologies. The braiding technique is the most suitable technique to produce structures with good dimensional stability and mechanical strength and high shape recovery and flexibility.^4,5 Normally, a braid (denoted biaxial braid) is constituted of two groups of yarns intertwined. The braiding angle, defined as the angle between the yarns and the longitudinal axis of the braids, is the main parameter in the kinematic analysis^6,7 of the braiding process due to its influence on the mechanical behavior of braided products.^8–10 For cardiovascular stents designed with biaxial braids, Rebelo et al.¹¹ thoroughly investigated the influence of braiding parameters (braiding angle, mono-filament type, braid diameter) on the mechanical behavior as well as the porosity of the stent. Potluri and Manan¹² focused on the geometric and micromechanical modeling of non-orthogonal structures to establish the parallel between braided structures and sheared woven structures. Harte and Fleck¹³ presented experimental results of the behavior of tensile of biaxial braids, and showed an increase of nominal axial strain in the function of the braiding angle. Hristov et al.¹⁴ and Dabiryan and Johari¹⁵ described the non-linear responses under tensile loads in the axial direction of biaxial braids. Del Rosso et al.¹⁶ compared the tensile behaviors of Dyneema®SK75 and Kevlar®49 biaxial microbraids.

It is necessary to have a high volume fraction of fibers for mechanical applications, which can be reached with triaxial braids where axial yarns are added along the longitudinal axis of the braids. Rawal et al.^17,18 described the tensile behavior of biaxial and triaxial braided structures and compared it with analytic models developed by various researchers. Relative to these studies, a few papers have experimentally studied the influence of other braiding parameters on the tensile behavior, such as the number of yarns. Moreover, the development of models^17,18 to optimize these braiding parameters requires experimental data on tensile behavior, and especially on the evolution of the braiding angle, its influence on the tensile behavior and the presence of a “double-peak,” as shown by Duchamp et al.^19,20 The evolution of the braiding angle due to the reorientation of bias yarns is the key parameter to understand the mechanism during the tensile tests of braided reinforcements. This phenomenon can be linked with the in-plane shear behavior of woven reinforcements,²¹ which is intensively described in the literature due to its influence on wrinkling and other defects during the composite manufacturing process.²² In addition, inter-yarn contact and friction are also related to the evolution of the angle between yarns. At the mesoscopic scale, friction between yarns was experimentally described taking into account the twist or angle between yarns.²³ As UHMWPE material is characterized by high thermal conductivity,^1,24,25 the idea proposed in this study is to measure the temperature of samples during tensile tests in order to associate the yarn/yarn friction with the evolution of the braiding angle, at the scale of braids.

In the present study, different samples of triaxial braids made from UHMWPE multifilament are designed with a large range of braiding angles and with changing numbers of axial yarns. In order to analyze the influence of these process parameters on the tensile behavior of the braided reinforcements, temperature and braiding angle measurements during tensile tests are carried out. Moreover, identification of the thermal behavior focused on the tensile behavior of these dry braids is conducted. A comparison of the thermal behaviors among various braiding angles is also illustrated experimentally.

Materials and methods

A multifilament UHMWPE yarn is used (density of 0.97 g/cm³) in these tests and its specifications and mechanical properties are given in Table 1. This characterization is necessary to regulate the tension during the braiding process.

Table 1.

Properties of ultra-high-molecular-weight polyethylene¹⁹

Properties	Value
Linear density (tex)	169
Strain at break (%)	3.9
Tenacity (cN/tex)	265
Twist (tr/m)	0

Braiding is a textile process that manufactures fabrics in flat or tubular form by intertwining three or more yarns together. In a braiding machine, the gear train is typically circular, consisting of specialized gears, called horn gears, composed of a spur gear bottom and a slotted gear top.^26,27 These gears move two sets of bobbin carriers in opposite directions so that the yarns interlace to design braids. The motion of the bobbins is accomplished by horn gears. From this principle, the braiding angle (denoted α) is a function of the process parameters. From this multifilament, triaxial braids (diamond pattern) are realized with a braiding loom, as described in previous works.^19,20,28,29 Figure 1 illustrates the scheme (without crimp) of a triaxial configuration.

Figure 1.

Scheme of triaxial braids.

The maximum number of carriers of the braiding loom used is 96 for bias yarns and 48 for axial yarns. For this study, three UHMWPE multifilaments were used per bobbin of bias yarns and, consequently, the number of bias yarns is constant and equal to 288. For axial yarns, one, two and three UHMWPE multifilaments were used per bobbin to consider respectively 48, 96 or 144 axial yarns. In our denomination of braid samples (Table 2), the first two numbers are respectively related to these numbers of bias yarns (first number) and axial yarns (second number) and they are expressed by the ratio of 48. The third number is related to the braiding angle. For example, a sample denoted 6-3-48 is a braid with 288 (6 × 48) bias yarns, 144 (3 × 48) axial yarns and with a braiding angle of 48°. The large range of braiding angles chosen is separated into three groups, denoted as small, middle and large, respectively. Table 2 shows detailed information of the samples.

Table 2.

The denomination of braids

Name of sample	Group
6-1-12 6-2-12 6-3-12	Small
6-1-20 6-1-29 6-2-21 6-3-23	Middle
6-2-42 6-2-68 6-3-48 6-3-54	Large

The braiding angles are not only measured during the braiding process but also when braids are removed from the braiding machine. These measurements are realized by optical techniques. Pictures of braids are taken by a camera and these pictures are analyzed using ImageJ software to compute an average of values at different places along the width and the length of the sample. A statistic study is made to ensure the accuracy of measurements on the same braided sample (precision ± 0.55°).

In addition, geometrical property (braiding angles) uniaxial tests on dry braids are conducted on an Instron tensile machine with a 250 kN load cell. According to standard NF ISO 13 934, the test speed is 20 mm/min (Figure 2). The ends of the sample are maintained by jaws realized with epoxy resin and unidirectional (UD) glass in order to avoid sliding under the grips of the tensile machine. All the tensile tests are conducted in the axial direction. The yarns are maintained between the grips of the device and, consequently, they are subjected to the tensile load. Axial yarns are oriented in the tensile direction, while bias yarns present one angle to the tensile load at the beginning of the test as the initial braiding angle. The results are obtained from the analysis of 10 samples tested for each configuration. The evolution of the braiding angle is monitored by a camera positioned in front of the machine during tests. Furthermore, a thermal camera measures the temperature during tensile tests by pointing at the sample. The viewable picture is shown in Figure 3. The temperature is measured in the center of the sample. Toward the end of the test, the hottest point is usually no longer in the center of the braid. The value of the highest temperature in the frame is accepted. The measures are precise to ±0.1℃.

Figure 2.

Test machine with the sample.

Figure 3.

Thermal behavior conducted by the thermal camera during the test.

Results and discussion

Tensile behavior

The load/deformation curves with error bars associated with the samples are presented in Figure 4. These results are clearly presented for the three groups depending on the range of the braiding angle, as discussed above (and denoted by “large,” “middle,” “small,” as defined in Table 2). Firstly, the samples with larger braiding angles are characterized by high tensile deformation contributing to high energy absorption, which could explain their application in ballistic protection.² Secondly, the tensile behavior of triaxial braids is characterized by a “double-peak” phenomenon. As explained by Duchamp et al.,¹⁹ during the test axial yarns aligned in the load direction are directly subjected to the tensile load. The first peak is related to the break of axial yarns and the second part of the curve is associated with the behavior of the bias yarns. Depending on the braiding angle value, the second part of the curve can be characterized by a non-linear zone associated with the rotation movement of the bias yarns, as described in detail for the in-plane shear behavior of wovens.²¹ At the second peak, all the yarns are broken.

Figure 4.

The tensile behavior for three groups of braid samples: (a) large; (b) middle; (c) small.

Due to the range of braiding angles, however, the double-peak phenomenon was different. For large braiding angles (Figure 4(a)), the first peak is reached at around 5% of strain, but the second peak is reached slowly and approximately at 150% of strain (6-2-28 sample). In the smaller braiding angle group (middle or small), the whole deformation decreases obviously and the first peak has more deformation than the second peak. Particularly for the small braiding angle (12°), the two peaks become closer. This can be explained by the fact that axial and bias yarns are both subjected to the load at the same time, since the braiding angle is so small that bias yarns are close enough to bear the tension, leading to the larger force in samples with a small braiding angle. The influence of braiding parameters (braiding angle and number of axial yarns) on the maximum deformation and load can be further analyzed in Figures 5 and 6, respectively. Maximum deformation values from these tensile tests are presented in Figure 5. The exponential increase of the maximal deformation in function of the initial braiding angle is illustrated independently from the number of axial yarns. In contrast, the initial braiding angle in the function of the load at the second peak shows a significant decreasing trend in Figure 6.

Figure 5.

Influence of the initial braiding angle on the maximal deformation.

Figure 6.

Influence of the initial braiding angle on the maximal load at the second peak.

The influence of the number of axial yarns on values of the maximal load at the first peak is shown in Figure 7. The increasing load values with the increasing number of axial yarns is significant, and can be illustrated by samples with almost the same braiding angles. For samples with an angle of 12° (or around 20°), a ratio of 2 in the number of axial yarns approximately involves a 20% increase of the load. For a constant number of axial yarns in the braids, it can be inferred that the maximal load at the first peak increases with the decreasing braiding angle. Therefore, the first part of the tensile behavior (first peak) cannot be associated with only the contribution of the axial yarns in the direction of the load applied. In Duchamp et al.,²⁰ contributions of the bias yarns have been proposed by models defined in TexMind Braider software.

Figure 7.

Influence of the number of the axial yarns on the maximal load at the first peak.

Evolution of the braiding angle during tensile tests

Due to the measurements during tensile tests, the results of the evolution of the braiding angle in the function of the deformation with error bars are as presented in Figure 8. The results are divided naturally according to the three groups of the initial braiding angle. In each group, the final value reached by the braiding angle at the end of the tensile test is clearly noted. It is noticeable that all the samples have an identical final braiding angle between 8° and 13° independently from their initial braiding angle. This final angle between bias yarns in the load direction is related to the width of the material used (linear density, Table 1) and the available space. From the experimental results, it could be expected that the evolution of larger braiding angles accompanies high deformation, which is inverse to the trend presented by smaller braiding angles (clearly seen in Figure 8).

Figure 8.

The evolution of braiding angle during tensile tests: (a) large; (b) middle; (c) small.

The evolution of the middle braiding angle can be clearly divided into two parts, as shown in Figure 8(b). In the first part, the evolution of the braiding angle decreases with a smaller slope before reaching the first peak of the load, since a small portion of the load is assigned to bias yarns. In the second part, however, the evolution has a relatively great decreasing trend with a higher slope until the final value. The first part also should be clearly shown in the large braiding angle, but the deformation at the first peak is too small to be observed directly compared to the whole high deformation in Figure 8(a). Comparatively, the evolution of the small braiding angle is different from that of the larger group since this decreasing trend seems to be linear. This could further confirm that axial yarns are the primary part sustaining the tensile force, while the bias yarns also bear load to a relatively small extent. This small extent would influence the evolution of the braiding angle greatly if the triaxial braids were designed with a small braiding angle. Hence, the linearly decreasing trend of the braiding angle could be seen at the beginning of the test (see Figure 8(c)).

Thermal characteristics during tensile tests

The evolutions of temperature in samples during tensile tests for different initial braiding angles with error bars are presented in Figure 9. It is clearly observed that the temperature increased drastically during the test. This phenomenon is related to the friction between yarns. Figure 10 shows the maximum temperature in the function of the number of axial yarns. Regarding the samples with the quasi-identical initial braiding angle (6-1-12; 6-2-12; 6-3-12 or 6-1-20; 6-2-21; 6-3-23), the maximum temperature increases following the increasing of the number of axial yarns. The increasing of the number of axial yarns leads to an augmentation of the intra-yarn fiction and, consequently, an augmentation of the maximum temperature can be observed during the tensile test. The maximal temperature reached by samples with middle or small braiding angles is higher than that with the large initial braiding angle when the number of axial yarns is fixed. This is probably because the axial and bias yarns simultaneously bear the load, which can directly increase the local friction, and the evolution of temperature at smaller braiding angles has inefficient space and time to dissipate the heat that increases the local temperature. Thus, the temperature is inversely proportional to the initial braiding angle.

Figure 9.

The temperature evolution with different initial braiding angles during tests: (a) large; (b) middle; (c) small.

Figure 10.

Maximum temperature in the function of the number of axial yarns.

To further analyze the thermal characteristics during the test, the relationship between strains, the evolution of load and the temperature for two samples of the middle and large groups are shown in Figures 11 and 12. As presented previously, the average curves were obtained from 10 tests. For the sample with the middle angle, it can be deduced that the increase of the temperature is maintained to a relatively small degree during the first peak of load evolution (about 4°). After the first peak, the temperature increases rapidly and almost follows the trend of load evolution to reach the second peak.

Figure 11.

Temperature and load evolutions during tensile tests for sample 6-1-29.

Figure 12.

Temperature and load evolutions during tensile tests for sample 6-3-48.

For braid 6-3-48 (in Figure 12), however, the temperature increases rapidly until reaching the first peak due to the number of axial yarns, which is higher in this case. However, the long time associated with the rotation of the bias yarns is characterized by a slightly decreased temperature (10–25% deformation). It is possibly because the friction caused by yarn rotation is too small to increase the temperature. As soon as bias yarns become contacted, which induces a significant increase in the load evolution, the temperature follows this increasing trend until the yarns are broken, since the friction is the dominant factor causing heat and, meanwhile, enough space to dissipate heat is scarce. Therefore, it is concluded that the number of yarns during the tensile stage has a great effect on increasing temperature because the friction between yarns under load increases with the increasing number of yarns. In addition, during the shear stage, the yarn rotation barely influences the temperature.

Figures 13 and 14 further explain the conclusion above from the point of view of the evolution of the braiding angle. It is clearly seen that the evolution of braiding angles and temperature have a definite inverse trend. For the 6-1-29 sample, when deformation exceeds 10%, the evolution of the braiding angle decreases slowly, and then this trend becomes gentle until the final value is reached. At this stage, the temperature increases more rapidly than it does before 10% deformation. Thus, it is suggested that temperature caused by friction rises quickly when bias yarns become contacted such that the “lock angle” is reached. Comparatively, in the 6-3-48 sample, the evolution of the braiding angle decreases greatly between 10% and 30% deformation, which is the stage when bias yarns bear the shear deformation. The temperature decreases slightly rather than increases; clearly, the friction by shear deformation of bias yarns is not enough to increase the temperature during the test, This is attributed to efficient dissipation resulting in enough time and space during the evolution of the braiding angle. Hence, the friction caused by yarns under the tensile stage strongly influences the temperature, while the friction caused by bias yarn rotation under shear behavior has little effect on temperature, especially for braids with a large braiding angle.

Figure 13.

Evolutions of temperature and braiding angle during tensile tests for sample 6-1-29.

Figure 14.

Evolutions of temperature and braiding angle during tensile tests for sample 6-3-48.

Conclusions

Fabrics made by UHMWPE multifilaments are novel in being considered as the reinforcement for natural fiber composites. In order to further improve the understanding of such reinforcements, their mechanical and thermal behaviors are investigated in depth in this paper. Generally, the triaxial braids show the “double-peak” phenomenon in load/deformation behavior. The maximum load at the “first peak” increases as the braiding angle decreases and the number of axial yarns increases. Moreover, the “first peak” is associated not only with the contributions of axial yarns but also with the contributions of bias yarns. This phenomenon is more obvious as the braiding angle becomes smaller. Triaxial braids with a larger braiding angle present the capacity of high deformation but their strength is weak. On the contrary, samples with a small braiding angle show the inverse mechanical behavior.

Regarding the evolution of the braiding angle, the triaxial braids with large and middle braiding angles have two distinctive decreasing parts: a small decreasing trend with the smaller slope and a large decreasing trend with higher slope. However, samples with a small braiding angle show an almost linearly decreasing trend, as the braiding angle could continually decrease from the beginning of test. This phenomenon further confirmed that the axial yarns and bias yarns are both subjected to the load, especially for the samples with a small braiding angle.

The temperature variation during the tests is also measured and analyzed. The temperature is primarily influenced by the intra-yarn friction, which is associated with design parameters including the number of axial yarns and the value of the braiding angle. Through the experimental analysis, it is clearly seen that the temperature increases with the increasing number of axial yarns due to incremental friction between yarns. As the axial and bias yarns sustain the load simultaneously, it leads to a strong increasing of the inter-yarn friction. Consequently, the maximum temperature can be reached with a small braiding angle. Moreover, the friction influences temperature severely at the stage of tensile behavior, while the yarn rotation under shear behavior has little effect on temperature, especially for braids with a large braiding angle. Therefore, it is necessary to select the braiding parameters, including the number of axial yarns and braiding angle, properly in order to control the temperature in a reasonable scope during the workpiece service.

Footnotes

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the China Scholarship Council.

References

Forster

Chin

Peng

J-S

, et al. Long term stability of UHMWPE fibers, Cham: Springer, 2016, pp. 369–375.

Zhou

Chen

Wells

. Influence of yarn gripping on the ballistic performance of woven fabrics from ultra-high molecular weight polyethylene fibre. Compos Part B Eng 2014; 62: 198–204.

Jahan

Walters

Riahinasab

, et al. A comparative study of radiation effects in medical-grade polymers: UHMWPE, PCU and PEEK. Radiat Phys Chem 2016; 118: 96–101.

Mouritz

Bannister

Falzon

, et al. Review of applications for advanced three-dimensional fibre textile composites. Compos Part A Appl Sci Manuf 1999; 30: 1445–1461.

Kyosev

. Generalized geometric modeling of tubular and flat braided structures with arbitrary floating length and multiple filaments. Text Res J 2016; 86: 1270–1279.

Potluri

Rawal

Rivaldi

, et al. Geometrical modelling and control of a triaxial braiding machine for producing 3D preforms. Compos Part A Appl Sci Manuf 2003; 34: 481–492.

Popper

. Analysis of a circular braiding process for complex shapes. J Text Inst 1994; 85: 316–337.

Ayranci

Carey

. 2D braided composites: a review for stiffness critical applications. Compos Struct 2008; 85: 43–58.

Ayranci

Carey

. Predicting the longitudinal elastic modulus of braided tubular composites using a curved unit-cell geometry. Compos Part B Eng 2010; 41: 229–235.

10.

Kier

Salvi

Theis

, et al. Estimating mechanical properties of 2D triaxially braided textile composites based on microstructure properties. Compos Part B Eng 2015; 68: 288–299.

11.

Rebelo

Vila

Fangueiro

, et al. Influence of design parameters on the mechanical behavior and porosity of braided fibrous stents. Mater Des 2015; 86: 237–247.

12.

Potluri

Manan

. Mechanics of non-orthogonally interlaced textile composites. Compos Part A Appl Sci Manuf 2007; 38: 1216–1226.

13.

Harte

Fleck

. On the mechanics of braided composites in tension. Eur J Mech A Solids 2000; 19: 259–275.

14.

Hristov

Armstrong-Carroll

Dunn

, et al. Mechanical behavior of circular hybrid braids under tensile loads. Text Res J 2004; 74: 20–26.

15.

Dabiryan

Johari

. Analysis of the tensile behavior of tubular braids using energy method, part I: theoretical analysis. J Text Inst 2016; 107: 553–561.

16.

Del Rosso

Iannucci

Curtis

. Experimental investigation of the mechanical properties of dry microbraids and microbraid reinforced polymer composites. Compos Struct 2015; 125: 509–519.

17.

Rawal

Saraswat

Kumar

. Tensile response of tubular braids with an elastic core. Compos Part A Appl Sci Manuf 2013; 47: 150–155.

18.

Rawal

Saraswat

Sibal

. Tensile response of braided structures: a review. Text Res J 2015; 85: 2083–2096.

19.

Duchamp

Legrand

Soulat

. The tensile behaviour of biaxial and triaxial braided fabrics. J Ind Text 2016; 47: 1–21.

20.

Duchamp

Kyosev

Legrand

, et al. Experimental and numerical investigation of triaxial braid reinforcements. Narrow smart text, Cham: Springer International Publishing, 2018, pp. 123–140.

21.

Boisse

Hamila

Guzman-Maldonado

, et al. The bias-extension test for the analysis of in-plane shear properties of textile composite reinforcements and prepregs: a review. Int J Mater Form 2017; 10: 473–492.

22.

Boisse

Colmars

Hamila

, et al. Bending and wrinkling of composite fiber preforms and prepregs. A review and new developments in the draping simulations. Compos Part B Eng 2018; 141: 234–249.

23.

Gupta

Chang

. Structural studies in continuous-filament yarns Part III: studies with fiber friction as a variable. Text Res J 1976; 46: 90–99.

24.

Zhou

Chen

. Numerical investigations into the response of fabrics subjected to ballistic impact. J Ind Text 2016; 45: 1530–1547.

25.

Guoliang

Hong

, et al. Influencing factors and measuring method of the heat conducting performance of UHMWPE single fiber. J Ind Text 2018; 47: 1908–1924.

26.

Nicolais L. Wiley encyclopedia of composites. Hoboken, NJ: John Wiley & Sons, Inc., 2011.

27.

Branscomb

Beale

Broughton

. New directions in braiding. J Eng Fibres Fabr 2013; 8: 11–24.

28.

Jacquot

Wang

Soulat

, et al. Analysis of the preforming behaviour of the braided and woven flax/polyamide fabrics. J Ind Text 2016; 46: 698–718.

29.

Risicato

Kelly

Soulat

, et al. A complex shaped reinforced thermoplastic composite part made of commingled yarns with integrated sensor. Appl Compos Mater 2015; 22: 81–98.