Effect of specimen thickness on high strain rate properties of kevlarpolypropylene composite

Effect of Specimen Thickness on High Strain Rate Properties of Kevlar/Polypropylene Composite Procedia Engineering 173 ( 2017 ) 694 – 701 1877 7058 © 2017 The Authors Published by Elsevier Ltd This is[.]

Procedia Engineering 173 ( 2017 ) 694 – 701

1877-7058 © 2017 The Authors Published by Elsevier Ltd This is an open access article under the CC BY-NC-ND license

( http://creativecommons.org/licenses/by-nc-nd/4.0/ ).

Peer-review under responsibility of the organizing committee of Implast 2016

doi: 10.1016/j.proeng.2016.12.151

ScienceDirect

11th International Symposium on Plasticity and Impact Mechanics, Implast 2016

Effect of Specimen Thickness on High Strain Rate Properties of

Kevlar/Polypropylene Composite

Hemant Chouhana,b,*, Neelanchali Asijaa, Shishay Amare Gebremeskela,

Naresh Bhatnagara

a Department of Mechanical Engineering, Indian Institute of Technology Delhi, New Delhi, 110016, India

b Department of Mechanical Engineering, Amity University Uttar Pradesh, 201313, India

Abstract

Characterization of Kevlar-Polypropylene based composite material system under high strain rate loading has been investigated using Split Hopkinson Pressure Bar (SHPB) test for varying specimen aspect ratios Flat laminates of 16, 24 and 30 layered Kevlar composite were compression molded and laser machining to get cylindrical specimens of desired aspect ratios Based on SHPB experiments, stress-strain plots were obtained and analysed to reveal compressive material behaviour as function of growing strain rate The peak stress, strain and toughness exhibited considerable increase with growing strain rate of loading With increasing strain rates peak specimen stress increased by 90%, for lowest thickness composite The aspect ratio studies suggests application of thin laminates for better performance of composite laminates

© 2016 The Authors Published by Elsevier Ltd

Peer-review under responsibility of the organizing committee of Implast 2016

Keywords: Kevlar; Thermoplastic matrix; High strain rate; Aspect ratio

1 Introduction

The need for light weight body armor had always been in demand This resulted in phenomenal growth of fiber reinforced plastic composites in last five decades The high tenacity fibers coupled with low specific gravity matrix

* Corresponding author Tel.: +91-011-26591139; fax: +91-011-26582053

E-mail address: hemant.chouhan78@gmail.com

© 2017 The Authors Published by Elsevier Ltd This is an open access article under the CC BY-NC-ND license

( http://creativecommons.org/licenses/by-nc-nd/4.0/ ).

Peer-review under responsibility of the organizing committee of Implast 2016

results into a composite system perfectly suited for the ballistic applications With ever increasing demand, the need

to study the material behaviour under high strain rate loading becomes essential The quasi-static testing done on universal testing machines cannot reveal the materials response under high strain rate of loading Therefore, there was a strong need of a testing scheme to study the material behaviour under dynamic high strain rates of loading This need was fulfilled in the form of a unique test facility developed by Kolsky [1], named as split Hopkinson pressure bar test apparatus

It has been reported in literature that most of the material whether metallic or non-metallic have strain rate dependent properties Allazadeh et al [2], studied steel, aluminum, wood and graphite-epoxy composite under high stain rate of loading and reported difference in rate dependent behaviour of all these materials Similarly, numerous studies revealed rate dependent behaviour of metals and composite Nolting et al [3] studied different grades of naval steel They optimized SHPB system to suit their materials and reported rate dependency of three grades of naval steel An increase in strength and peak strain was reported for different metals with increasing strain rate of loading [4][5]

The rate dependent behaviour of FRC’s has also been studied Song et al [6], studied S-2 glass/SC15 composite along thickness direction and in-plane direction under high rate loading on SHPB and the peak strength obtained were 700 MPa and 500 MPa for peak strain rate with in the experimental strain rate range Xuan et al [7], studied woven carbon/epoxy under dynamic loading and reported dynamic strength limit higher than quasi-static strength limit Woo et al [8][9], studied Kevlar composite under high strain rate loading using acoustic emission along with SHPB and reported higher strength as a function of growing strain rates of loading The peak stress acquired by Kevlar composite varied in range of 160 ~ 370 MPa for an experimental strain rate range of 1182 ~ 1460 s-1 Ramadhan et al [10], studied the effect of three different specimen thicknesses on dynamic properties of Kevlar/epoxy and Aluminum laminated panels under ballistic impact and reported change in material behaviour as a function of specimen thickness

Nomenclature

Hr reflected strain

Ht transmitted strain

C0 elastic wave velocity in the bars

E Young’s modulus of elasticity of bar material

AB cross-section area of the bar

AS cross-section area of the specimen

Ls specimen length

t time duration

X poisson’s ratio

FRC fiber reinforced composite

SHPB split Hopkinson pressure bar

K-PP Kevlar-Polypropylene

Kevlar composites have been used with thermosetting polymers for ballistic applications However, the need arises for using thermoplastic polymers as matrix to further reduce the weight of FRC armor A number of experimental works on varying specimen aspect ratios are available in literature Knowing that ideal aspect ratio resulting into negligible inertial effects is given as ξͲǤ͹ͷX [11] The objective of present study is to characterize three different thicknesses of Kevlar-thermoplastic composite system under high strain rate of loading The 16, 24 and 30 layered Kevlar-Polypropylene composite with constant diameter of cylindrical specimen results into three different aspect ratios (0.3, 0.42 and 0.53) were tested under high strain rate loading on SHPB Kevlar composite having 24 layers is close to theoretical requirement and other two will serve as lower and higher value of aspect ratio The test results are of importance due to the fact that whereas, SHPB test reveals the properties on the basis of one-dimensional loading, the physical loading while in service is three-dimensional in nature Hence, the test results

will help understand the effect of varying number of Kevlar mat layers on high strain rate loading of identical diameter composite specimen

2 Experimental Details

An experimental SHPB set-up was developed in-house for testing of various materials under high strain loading (Fig 1) The major sub-components of SHPB apparatus are gas gun, striker bar, incident bar, transmission bar, momentum trap, speed sensor, signal conditioner & amplifier and data acquisition system [12] Majority of works reported in literature used steel grades for testing of composites To keep the impedance mismatch within close range softer materials like Aluminum should be preferred However, lack of hardness results in deformation of Aluminum bars Therefore, Titanium bars having diameter of 16 mm were used Titanium is having lower density than steel and much harder than Aluminum The specifications of Titanium bars and strain gauges used for compressive SHPB development are presented in Table 1

Fig 1 Compressive Split Hopkinson Pressure Bar schematic arrangement

Table 1 Specifications of SHPB Set-up

Bar properties

Strain gauge properties

2.1 Operation of SHPB

The striker bar acquires velocity when the pressurised Nitrogen gas is released by operating solenoid value controlled by a PLC suitably programmed for the purpose The impact of striker bar onto the incident bar generates elastic compressive stress wave in the incident bar When the elastic stress wave crosses the strain gauge mounted at mid of incident bar the same is recorded in the form of voltage change generated due to change in resistance of strain gauge The stress wave then enters the specimen placed at the end of incident bar When the stress wave passes through the specimen, a part of stress wave is reflected back into the incident bar and rest of the wave passes through the specimen into the transmission bar The strain/deformation history of specimen is recorded in the form

of voltage signal generated due to change in resistance of strain gauge on incident and transmission bar The voltage

signal received from incident and transmission bar gives reflected and transmitted pulse voltage responsible for specimen strain, strain rate and stress induced in the specimen Quarter bridge arrangement based on Wheatstone bridge was used for recording the strain developed on the bars The strain gauges were powered by signal conditioner and amplifier and the data received in the form of voltage was stored by NI-6115 onto a dedicated PC From these measured voltage signals using one-dimensional wave propagation theory, the strain, strain rate and stress induced in the specimen can be assessed

2.2 Specimen preparation and design

Flat composite laminates of Kevlar-129 (Grade: 802F, Make: DuPont) with high impact Polypropylene (Grade: Repol CO15EG, Make: Reliance Polymers) were fabricated by compression molding The Kevlar mat was used as provided by the manufacturer The PP polymer was available in the form of granules By the process of film extrusion, the polymer was converted into film of 0.02mm thickness The bonding of PP with Kevlar is not good, therefore, 10% Malic Anhydride grafted Polypropylene (MAgPP) was blended to enhance the bonding between fiber and PP matrix Compression molding was done at 2000 C, 10 bar pressure under 500 mm of Hg vacuum To achieve three different aspect ratios of specimen the number of Kevlar layers were kept as 16, 24 and 30, respectively Laser machining was done on 400W continuous wave fiber laser (Model: RS400, SPI lasers, UK) to get cylindrical specimen of K-PP composite from flat composite laminates The laser machining was done perpendicular to flat surface of laminate The diameter of specimen were kept constant at 11.5 mm, resulting into theoretical aspect ratio of 0.3, 0.42 and 0.53 respectively, with in an experiment tolerances of ±5% error

Fiber volume fraction and density plays a significant role in deciding the mechanical properties of composites Desolution by solvent was performed to determine fiber volume fraction of composite laminates Xylene at 80 ˚C (atmospheric pressure) for 2 hours is used to dissolve polymeric matrix content of the K-PP composite specimen The fiber volume fraction and density determined are presented in Table 2

Table 2 Fiber volume fraction of Kevlar-Polypropylene laminates

No of Kevlar

Pulse shapers are recommended to modify the shape of incident pulse It has been reported to use a pulse shaper giving rise time identical to material response reported by transmission wave when experimented without a pulse shaper [13] Thus, Linatex a natural rubber was optimized to diameter of 3.1 mm and thickness of 1.4 mm to

be used as pulse shaper

3 High Strain Rate Testing

One-dimensional wave propagation in elastic bars with particle motion in longitudinal direction is the basis for development of SHPB Complete theory, instrumentation and methodology of SHPB is explained in literature [14][15] The derivations of SHPB are based on assumptions, which are prerequisite to achieve one-dimensional loading of the specimen That includes, homogenous state of stress in the specimen, negligible friction effects, negligible inertial effects and negligible wave dispersion along with perfect flat ends for contact with the bars However, due to continuous stress wave attenuation as a function of fiber-matrix deboning the condition of homogenous state of stress in the specimen is not successfully fulfilled in SHPB of composites [16][17] The analytical relations for determination of specimen strain, strain rates and stress as function of time are as follows:

The strain rate, ɂሶs(t) = ሺଶ஼బ

The average strain, Hs(t)= േሺଶେబ

The average stress, V(t) = േܧ ୅ా

During the SHPB testing, two methods of determining strain rate are prevailing In general, the strain rate of homogenous materials is determined by initial slope of strain rate-time curve However, due to stress wave attenuation there is substantial variation in reflected strain wave In this case, Nakai et al [18] suggested that strain rate can be calculated by dividing the area under the strain rate-strain curve up to maximum strain under loading by the maximum strain

3.1 Calibration of SHPB apparatus

Calibration of SHPB apparatus is essential to validate the conditions of one-dimensional wave theory and assess the accuracy of apparatus During calibration, both the incident and transmission bars are pressed together without a specimen Lubricant layer is applied in between the bars to minimize the friction In this condition both the bars can

be treated as single bar Experimentation was done in this state and resulting voltage signals on incident and transmission bar were recorded and analyzed Fig 2 (a) depicts voltage signals received on incident and transmission bar This voltage pulse is converted to equivalent force versus time curve for both incident and transmitted pulse (Fig 2(b)) Force history obtained for incident and transmitted pulse is F1 and F2 respectively Both the forces F1 and F2 matches well, indicating identical stress state in both the bars This ensures SHPB is perfectly aligned, friction free and ready for experimentation

Fig 2 Calibration results of compressive SHPB setup (a) strain gauge signals and (b) comparison of force vs time behaviour derived from strain

gauge signals

4 Results and Discussion

High strain rate compressive behaviour of 16, 24 and 30 layered Kevlar-PP composite is studied To calculate dynamic stress-strain properties, the data recorded in the form of incident, reflected and transmitted wave voltage is suitably fed to equations (1) to (3) Strain rates attained and resulting compressive properties for all the experiments

is presented in Table 3 The strain rate is defined at its peak and average for all the experiments in Table 3

For 16 layered K-PP composite, a uniformly rising stress with increasing strain rate is noted (Fig 3 (a)) A rising slope of stress-strain curve as function of strain rate is observed till a strain rate of 5442 s-1 Further higher striker bar velocity only resulted in damaged specimen Associated with damaged specimen was higher strain and strain rate due to increased area under stress-stain curve For strain rates below limiting value of 5442 s-1, a stress unloading curve was observed and the resulting permanent strain was noted below 0.2 After limiting strain rate, peak strain increased as a function of increasing strain rate, without much increase in peak stress The limiting strain rate, therefore refers to the strain rate at which the specimen just fails and there is neither a stress unloading curve nor a gradual stress fall A secondary loading curve was observed for further higher strain rate of loading at 9965 s-1 For 24 layered K-PP composite, a phenomenal change was observed in stress-strain behaviour for strain rates below limiting strain rate of 4048 s-1 A loop was formed at the peak of stress when strain starts dropping but stress continuous to grow (Fig 3 (b)) This phenomenon indicates growing strain signal on transmission bar as a function

-0.3

-0.2

-0.1

0

0.1

Time (s) (a)

Incident pulse (v) Transmitted pulse (v)

0

10

20

30

40

50

Time (s) (b)

F1 F2

of weakening strain signal on incident bar After limiting strain rate of loading, an insignificant growth in modulus is observed with marginally higher peak stress However, a significant growth in peak strain continues after limiting strain rate A secondary loading curve for 24 layered K-PP composite was noted at strain rate of 4379 s-1

For 30 layered K-PP composite, the growth pattern of stress was found different than the previous two cases (Fig

3 (c)) Whereas, a linear growth in plastic region was observed after insignificant stress growth in linear elastic region for 16 and 24 layered K-PP composite, a non-linear growth pattern in plastic region was observed in case of

30 layered K-PP composite The permanent strain for strain rates below limiting strain rate was in a narrow band After acquiring limiting strain rate, 30 layered K-PP composite also resulted in higher strain indicating consistency

of performance of K-PP composite

With increasing strain rate of loading, the shape of stress-strain curve at peak changes from a continuous curve to sharp peak followed by gradual fall The strain recovery is recorded maximum for 24 layered K-PP Indicating importance of ideal specimen aspect ratio After limiting strain rates of loading insignificant change is noted in stress as a function of rising strain rates It may be noted that for strain rates below limiting stress a significant change in both the slope of stress-strain curve and peak of stress is observed With increasing number of fiber layers, the position of initiation of secondary loading curve shifts on strain scale from 0.7 for 16 layered K-PP to 0.35 for

24 layered K-PP and 0.26 for 30 layered K-PP composite, respectively The peak stress attained by 16 layered K-PP composite was maximum at 753 MPa when loaded at limiting strain rate of 5442 s-1, whereas, higher number of layers resulted into lower peak stress and stain rate 738 MPa and 661 MPa at 4048 s-1 and 2837 s-1 for 24 and 30 layered K-PP composite, respectively

Fig 3 Effect of aspect ratio on dynamic compressive behaviour of K-PP composite compression molded at 10 bar (a) 16 layered K-PP

composite, (b) 24 layered K-PP composite and (c) 30 layered K-PP composite

Fig 4 depicts the aspect ratio effect on high strain rate behaviour of K-PP composite (average aspect ratio mentioned in bracket for respective number of Kevlar layers) The stress growth is presented till limiting strain rate

is achieved Since, after limiting strain rate specimen is failed and resulting strain growth increases strain rates For a given peak stress, thin specimen acquired highest strain rate of loading The K-PP specimen with minimum thickness resulted in highest strain rates of loading 30 layered K-PP composite resulted in lowest strain rates for identical peak stress value The 24 layered composite results lies in between the two extremes A linear curve fitting

0

200

400

600

800

Strain (a)

3307 s·¹

5442 s·¹

6708 s·¹

9965 s·¹

0

200

400

600

800

Strain (b)

2548 s·¹

4048 s·¹

4105 s·¹

4379 s·¹

0

200

400

600

800

Strain (c)

2101 s·¹

2837 s·¹

3481 s·¹

3515 s·¹

is adopted to reveal the nature of stress growth as a function of strain rate for different aspect ratios of specimen The linear curve fits well for K-PP composites upto limiting strain rates having different aspect ratios The coefficient of determination indicating closeness with the fitted regression line is above 0.99 for all the aspect ratios

Fig 4 Effect of aspect ratio on K-PP composite loaded up to limiting strain rates

Table 3: Strain rate effect on compressive properties of Kevlar-Polypropylene composites

No of Kevlar

layers

Fracture Modulus ‘F’

K-PP-16

K-PP-24

K-PP-30

5 Conclusion

To reveal the effect of aspect ratio on high strain rate behaviour of K-PP composite SHPB testing was done Following conclusions may be drawn on basis of high strain rate testing of 16, 24 and 30 layered K-PP composite

1 Fracture modulus grows as a function of strain rate of loading, till limiting strain rates of loading

2 Growth pattern of stress and strain were found linear in nature till limiting strain rate, but the toughness continues to follow growth even further at higher strain rates of loading, due to increasing specimen strain

as a function of strain rate

3 Thin specimen proved to be more advantageous over thick specimen, as far as peak stress is the criteria

4 A peculiar behaviour of loop formation at peak stress was observed in case of 24 layered K-PP composite

5 For a given striker bar velocity, varying strain rates were attained by the specimen of different thicknesses

6 Secondary loading of specimen was depicted in all cases, expect 30 layered K-PP composite

y = 0.1679x - 157.13 R² = 0.9979

y = 0.1741x - 26.733 R² = 0.9996

y = 0.3517x - 342.18 R² = 0.996

300

400

500

600

700

800

Strain rate (s-1)

K-PP-16 (0.3) K-PP-24 (0.42) K-PP-30 (0.53)

Acknowledgments

The authors are thankful to IRD-IITD for grand challenge (MI00810) for granting this research project

References

(1949) 676–700

Epoxy Composites in Response to High Strain Rate Load, Adv Mater Sci Appl 1 (2012) 13–30

(2011) 2268–2273

SHPB, Procedia Eng 24 (2011) 773–777

compressive loading, Compos Part B Eng 60 (2014) 531–536

B 97 (2016) 317–328

B Eng 60 (2014) 125–136

aluminum laminated panels, Mater Des 43 (2013) 307–321

Phys Solids 11 (1963) 155–179

Pressure Bar for Light and Soft Armour Materials, Glob Jr Res Eng XIV (2014) 62–75

Bar Testing, Metall Mater Trans A 38 A (2007) 2655–2665

(2011) 1567–1576

Compos Struct 52 (2001) 405–417

pressure bar, Int J Adhes Adhes 56 (2015) 13–23