ectc section 2 cover...composite beams”, i negru, g r gillich, z i praisach, m tufoi, n gillich....

UAB School of Engineering - Mechanical Engineering - Journal of the ECTC, Volume 18 Page 22

SECTION 2

Materials & Testing

UAB School of Engineering - Mechanical Engineering - Journal of the ECTC, Volume 18 3


Journal of UAB ECTC Volume 18, 2019

Department of Mechanical Engineering The University of Alabama, Birmingham

Birmingham, Alabama USA

STUDY OF THE SHIFT IN MATERIAL PROPERTIES DUE TO STRUCTURAL DAMAGE IN BIO-COMPOSITES

Gavin Stockstill, Maharshi J. Dave, Tejas Pandya University of Mississippi Oxford, Mississippi USA

Jason Street Mississippi State University

Mississippi State, Mississippi USA

ABSTRACT The natural frequency responses of three wood-based bio-

composites with varying volume fractions of methylene diphenyl diisocyanate (MDI), microcrystalline cellulose (MCC), and processing times and pressures have been studied through experimental modal analysis. The samples used for the study consisted of southern yellow pine particles with 1% MCC 4% MDI; - 2% MCC 4%MDI; - and 4% MDI compressed at 10.5 MPa (with ram pressure of 27.58 MPa). The dynamic and damping properties were obtained using the hammer excitation vibration technique. The natural frequency responses were obtained from both frequency and time domains, and the results were consistent with an average percent difference of 0.45%. The natural frequency responses from both damaged and undamaged samples have been found to observe the shift in natural frequency caused by damage in the material. When damage was induced in the samples, the average shifts for the first natural frequencies were 2.07%, 2.12%, and 1.64% for the 4% MDI, 1% MCC, and 2% MCC samples respectively. KEYWORDS: Natural Frequency, Damage, Experimental Modal Analysis, Wood Based Bio-Composites, Hammer Excitation, Vibration

INTRODUCTION A natural frequency exists in every structure. Ideally, natural

frequencies will have no effect on the reliability and properties of the structures they exist in. When damage occurs, a shift in the natural frequency of a material or structure has been detected [1]. Extensive research has been done on this concept. I Negru et al. [2] researched the effect that transverse cracks have on natural frequency and stiffness in structures. Rangel et al. [3] studied natural frequency shift in structures and proposed methods to repair damaged bridges.

Vibrational behavior of composite materials has been studied many times [4-6]. Vibrational methods have been implemented for many studies involving the determination of the mechanical and physical properties of wood and wood-based bio-composites. Similarly, these methods have been used in the forestry industry [7-9]. Zeng Wanga et al. [10] used the beam vibration technique to determine the dynamic Young’s modulus

and damping ratio for three commercially available wood-based composites.

Many studies have been conducted on the prediction of a shift in natural frequency due to structural damage. This study uses experimental modal analysis using the hammer excitation vibration technique to compare the shift in natural frequency after damage is induced for three different bio-composites, each consisting of a different volume fraction of microcrystalline cellulose.

MATERIALS The specimens were created from Southern yellow pine with

a 2-3 mm particle size. The temperature used to form the panels was approximately 185°C. Except for the panel compressed at 10.5 MPa, which used 5.9 kg of mass, all the panels were created with approximately 2.95 kg of mass. The panels were created using a Dieffenbacher 915 x 915 mm hot press system located at the Sustainable Bioproducts Laboratory at Mississippi State University coupled with the Alberta Research Council’s Pressman operation and monitoring software. The Dieffenbacher hot press forms panels based on a given thickness, so each bio-composite material required different pressures to produce a panel with a thickness of 6.35 mm. The wood-based bio-composite samples created for the analysis of the effect of structural damage on natural frequency using the hammer excitation vibration technique were made from microcrystalline cellulose (MCC) and Methylene Diphenyl Diisocyanate (MDI) resin with varying mass fractions. These can be found in Table 1.

MDI is an aromatic diisocyanate and is an efficient binder that has been used in the production of composite wood products for over 30 years. Microcrystalline cellulose was formed from techniques described in a previous study Chauhan et al. (2009) using pure cotton. The hammer excitation vibration test technique sample dimensions were 317.5 mm in length x 25.4 mm in width x 6.35 mm in height. For the tests with the damaged samples, damage in the form of a 0.25-inch hole was induced one inch from the fixed end.


Table 1: Constituent Materials for Wood-based Bio-Composites

Material Approx. Pressure (MPa)

Curing Time (seconds)

Density (kg/m3)

4% MDI 10.5 MPa (2x material was used)

10.5 140 1336.2

2% MCC 4% MDI

7.1 140 906.93

1% MCC 4% MDI

9.2 140 976.99

EXPERIMENTAL PROCEDURE AND SETUP The natural frequencies of both the damaged and

undamaged samples were found through experimental modal analysis using the hammer excitation vibration technique. The setup is in the Structure and Dynamics Lab at the University of Mississippi. Figure 1 shows the schematic.

Figure 1: Schematic of experimental setup

Using an impact hammer, the beam was excited. An

accelerometer attached to the free end of the beam detected the response. The accelerometer was attached to both an oscilloscope and a signal analyzer with a conditioning amplifier. The impact hammer was also connected to the signal analyzer through the conditioning amplifier. From the dynamic response initiated by the impact hammer, time dependent responses were amplified through power amplifiers and converted into a frequency response function (FRF) using the digital signal analyzer (Hewlett Packard, Model #35665A). The digital signal analyzer provided the imaginary component of the FRF magnitude, of which the first three peaks were plotted to form three different mode shapes at the respective natural frequencies.

EXPERIMENTAL RESULTS AND ANALYSIS

Using the cantilever beam vibration technique, the dynamic Young’s modulus was obtained for each sample before and after damage was induced. These can be found below in Table 2.

Tale. 2 Dynamic Young’s Modulus for Wood-Based Bio-

composites

Material E (Undamaged)

GPa

E (Damaged)

GPa

% Difference

4% MDI 4.7326 4.5852 3.16

1% MCC 4% MDI

3.3878 3.2508 4.13

2% MCC 4%MDI

2.4449 2.3645 3.34

Table 2 shows that the dynamic Young’s modulus decreased

when damage was induced on each sample. Also, it shows that a higher concentration of microcrystalline cellulose and a lower pressure in the creation of the samples resulted in a lower dynamic Young’s modulus for the bio-composite samples. However, the largest shift occurred in the material consisting of 1% volume fraction of MCC. Due to the high forming pressure of the 4% MDI sample, the mass density was increased resulting in a higher Young’s modulus and a smaller shift due to damage.

Similarly, inducing damage in the bio-composite materials resulted in a shift in their respective natural frequencies. Table 3 shows the average natural frequencies as well as the percent difference between the damaged and undamaged samples.

Table 3 Shift in Natural Frequency for Wood-Based Bio-

Composites

Material Natural Frequency

(Undamaged) Hz

Natural Frequency (Damaged)

Hz

% Difference

4% MDI 43.8125 43 1.87

1% MCC 4% MDI

39.375 38.5625 2.09

2% MCC 4%MDI

34.5 33.9375 1.64

The largest shift in both natural frequency and Young’s

modulus occurred in the material with 1% volume fraction of MCC. The smallest shift in both natural frequency and Young’s modulus occurred in the material with 2% MCC. This data implies that while the increased volume fraction of MCC at lower forming pressures can significantly lower the Young’s modulus, it is effective in reducing the effects that damage has on natural frequency and Young’s modulus.


CONCLUSION This study compared the material behavior in three different

bio-composites with different volume fractions of microcrystalline cellulose (MCC) and forming pressures, before and after damage was induced, using Young’s modulus and natural frequency. Based on the results and analysis listed above, it can be said that while a lower forming pressure and higher volume fraction of MCC lower both the natural frequency and the Young’s modulus, a MCC volume fraction of 2% and above can reduce the effects of structural damage on material behavior. When damage was induced in the three bio-composites, the material with the highest volume fraction of MCC showed the lowest shift in natural frequency at 1.64%. The material formed at the highest pressure (4% MDI formed at 10.5 MPa) showed the lowest shift in Young’s modulus, 3.16%, due to the high mass density resulting from the forming pressure.

ACKNOWLEDGEMENTS I would like to thank Dr. P. Raju Mantena for the lab

equipment, Dr. Tejas Pandya for the assistance in completing the research, Maharshi Dave and Damian Stoddard for help with the experiments, and Matthew Lowe for help with the preparation/machining of the specimens.

REFERENCES [1] Aarsh Shah, Tejas S Pandya, Damian Stoddard, Suman Babu Ukyam, Jason Street, James Wooten, Brian Mitchell; “Dynamic response of wood-based bio-composites under high-strain rate compressive loading.” Journal of Wood and Fiber Science, 2017. [2] “Natural frequency changes due to damage in composite beams”, I Negru, G R Gillich, Z I Praisach, M Tufoi, N Gillich. 11th International Conference on Damage Assessment of Structures (DAMAS 2015), IOP Publishing, Journal of Physics: Conference Series 628 (2015) 012091 doi:10.1088/1742-6596/628/1/012091 [3] Jamie Horta-Rangel, Socorro Carmona, Victor M. Castano, (2008) “Shift of natural frequencies in earthquake damaged structures: an optimization approach”, Structural Survey, Vol. 26 Issue: 5, pp. 400-410 [4] Assessment of theories for free vibration analysis of homogenous and multilayered plates Shock Vib. 11(3-4) 261-70 [5] Hause T and Librescu L 2006 Flexural free vibration of sandwich flat panels with laminated anisotropic face sheets J. Sound Vib. 9(4-5) 607125 [6] Gillich G R, Praisach Z I, Wahab M A and Vasile O 2014 “ Localization of transversal cracks in sandwich beams and evaluation of their severity” Shock Vib. 9(4-5) 607125 [7] Moslemi AA. Dynamic viscoelasticity in hardboard. Forest Prod J 1967;17(1):25-33 [8] Ross Rj, Pellerin RF. Nondestructive testing for assessing wood members in structures: a review. USDA Forest service Forest Products Laboratory, General tech. rep. FPL-GTR-70; 1994 [9] Ilic J. Dynamic MOE of 55 species using small wood beams. Holz Roh Werkst, 2003;61(3):167-72

[10] Measurement of dynamic modulus of elasticity and damping ratio of wood-based biocomposites using the cantilever beam vibration technique. Zheng Wanga, Ling Li, Meng Gong. Journal of Construction and building materials. doi:10.1016/j.conbuildmat.2011.09.001





DYNAMIC COMPRESSIVE BEHAVIOR AND ENERGY ABSORPTION OF ETHYLENE POLYURETHANE DIENE MONOMER FOAM USING SPLIT HOPKINSON PRESSURE BAR

TECHNIQUE

Rowan E. Baird University of Mississippi Oxford, Mississippi USA

Mason Tudor University of Mississippi Oxford, Mississippi USA

Mr. Damian Stoddard

University of Mississippi Oxford, Mississippi USA

ABSTRACT The purpose of this study was to examine the dynamic

response of three different low impedance Polyurethane Diene Monomer Foams. The three variations of density were tested using the Split Hopkinson Pressure Bar (SHPB) and Digital Image Correlation (DIC) technique at three different strain-rates, 1800/s, 2400/s, and 2800/s. The results showed that the medium softness (20 Shore O) had the lowest and highest energy absorption of 1.231 KJ/Kg and 2.548 KJ/Kg at 1800/s and 2800/s respectively. It was also observed that the medium softness polyurethane showed the most sensitivity to strain-rate within the tested parameters. Medium softness polyurethane foam was found to have the highest compressive strength at a strain-rate of 2800/s. Key Words: Ethylene Propylene Diene Monomer (EPDM) Foam, Split Hopkinson Pressure Bar (SHPB), Energy Absorption, High Strain-Rate, Dynamic Compression.

INTRODUCTION The Ethylene Propylene Diene Monomer (EDPM) Foams

are available in a variety of thicknesses, sizes, colors and stiffness to meet application-specific performance requirements. Many could be used for packaging, shipping, or shoe insoles. To understand the full response of these foams, a variety of experiments needs to be completed. One of these is understanding the responses to a variety of strain-rates. To test at high strain-rates, scientists often use the Split Hopkinson Pressure Bar (SHPB) technique. The split Hopkinson pressure bar has been used widely and modified to determine dynamic properties of a variety of engineering materials, but soft materials pose a problem [1]. During dynamic characterization, the mechanical impedance is extremely low and noise level typically prevents proper transmitted signals to be obtained from the conventional experiment. In this paper, results of an

extremely soft foam are analyzed using the modified Split Hopkinson Pressure Bar (SHPB) and Digital Image Correlation (DIC) technique for characterizing at high strain-rates [1]. The material’s strain-rate sensitivity is also discussed [2]. Ouellet saw strain-rate sensitivity for expanded Polystyrene and high-density polyethylene and rigid polyurethane. Increase in stress in EPS and HDP for different densities was observed, as well as an increase in lower density (PU) and a decrease in compressive strength with increased strain-rate [3]. Tedesco et al. also found strain-rate sensitivity for phenolic foam and rigid polyurethane foam. An increase in stress-strain response was observed with an increase in polyurethane density. They also observed densification at approximately 30-40% compressive strain [4]. Whisler et al. saw a strain-rate hardening effect for polyurethane foam with increase in strain-rate. They also saw densification in the material with higher compression of material [5].

MATERIALS AND METHODS All three foams were obtained from McMaster-Carr, a

private supplier of raw materials, tools and equipment. The foams arrived in 30.5x30.5 cm2 sheets with 0.635 inch thicknesses. Foam samples were then cut into a circle with a diameter of 11.5 cm by a Full Spectrum 90-watt Laser 16x24 in the Machine Shop at the University of Mississippi. Durometer readings for each sample were obtained. The dynamic events were video recorded using a Shimadzu HPV- 2 High-Speed Video Camera. Due to the high frame rate, high levels of illumination were required. The required illumination was provided by Photogenic PLR2500DR strobes. Sample voltages were obtained from the data acquisition system’s strain gauges on the transmission bar, which were used to obtain the stress in the sample using Equation 1. Sample strain was obtained by ProAnalyst, a DIC software using Equation 2. Equation 3 was used when calculating the strain-rate of the samples.


Equation 1: 𝜎 𝑡 𝐸𝜀 𝑡 [1]

Equation 2: 𝜀∆

[6]

Equation 3: 𝜀 [1]

Experimental Setup

Dynamic testing for characterization of foam samples at high strain-rates was carried out on a modified SHPB in the Blast and Impact Dynamics Lab at the University of Mississippi. Aluminum bars of 19.02 mm diameter were used as striker and incident bars, and a hollow aluminum bar with an outer diameter of 19.02 mm with a wall thickness of 1.5 mm (inner diameter of 17.52mm) was used for the transmission bar. The transmission bar had end caps of 12 mm thick that went 38 mm deep into the bar as seen in Figure 1. This provided a flat surface interaction between the bar and the samples with a strength to withstand repeated testing without failure. A pulse shaper made of copper was placed in between the striker and incident bars to ramp the incident pulse in order to achieve constant stress rate [7,1]. A Shimadzu HPV-2 High-Speed Video Camera with a fixed resolution of 312 x 260 pixels at a recording speed of 250,000 fps was used to capture the deformation/failure process. The illumination was provided by a strobe system to capture images and provide enough lighting during deformation due to minimum exposure time. Figure 2 shows the camera and strobes used for this experiment. Figure 3 shows the camera setup with the computer connection to view the video for DIC analysis. Figure 4 is a schematic of the SHPB system.

Figure 1: Hollow Bar End Cap Schematic

For compressive loading, EPDM foam specimens are placed between the incident and transmission bars as shown in Figure 5. Once the stress pulse reaches the sample, the foam was deformed due to impedance mismatch. When the deformation occurs, the Shimadzu HPV-2 High-Speed Video Camera captured the deformation for Digital Image Correlation (DIC) Analysis later on. The three softness variations were tested at three strain-rates, 1800/s, 2400/s, and 2800/s. Figure 5 depicts what the samples go through during the experiment.

Figure 2: Shimadzu HPV2 camera and PLR2500DR

strobes used during testing

Figure 3: shows camera recording setup

Figure 4: Split Hopkinson Pressure Bar Schematic [8]


Figure 5: Dynamic compressive loading on EPDM foam

specimen.

RESULTS The impact tests were performed using a split Hopkinson

pressure bar for three varying softness EPDM foams. Each material was tested at three different strain-rates, 1800/s, 2400/s, and 2800/s. For laboratory simplification, the three different levels of softness samples were painted three different colors. The nominal Shore O values of white, light blue, and dark blue samples are as follows 15.5 Shore O, 20.0 Shore O, and 25.0 Shore O, respectively. These values were measured via a Durometer O and from hence forth will be referred to by their color designation white, light blue, and dark blue respectively. Figure 6 shows a typical wave response from the material during the procedure, specifically this is Sample 13 (a dark blue sample). The response of the transmission bar appears to be nonexistent when in fact it is present just minimal in comparison to the incident bar’s responses. Figure 7 shows a graph of just the transmission wave for perspective. Figure 8 show a typical stress-strain response for EDPM foams. The material has the two notable characteristics, plateauing and densification, however, the responses are much softer than what is found for rigid foams. Plateauing is the phenomena when a foam responds fluidly to the compression force being applied during the SHPB testing, caused by a material plastically flowing due to an applied stress. Plateauing is identified on the stress-strain curve when the stress ceases to increase while the strain continues to increase creating a flattened area of data points that resembles a plateau. For this specific sample plateauing occurs between points 0.09 strain and 0.18 strain. Densification, a phenomenon that occurs due to compaction of a material which causes variations in density, can be seen for all foam types. Densification is identified when the stiffness of the stress-strain curve increases. In this specific sample’s response, densification occurs between 0.25 and 0.6 strain. Figure 9 is the video overlay of the stress-strain response of the material during the procedure. Utilizing a DIC technique, strain was obtained using video files. Initially, there is a rapid increase in stress between 0 and eight microseconds, reaching its plateau rather quickly. The ultimate strength is achieved at around 400 microseconds.

Figure 6: Typical Stress Wave Response for EDPM

Figure 7: Transmission Wave Response

Figure 8: Light Blue at 2400/s Strain-Rate

Figure 9: DIC Video Matchup to Numerical Data

‐2.1

‐1.6

‐1.1

‐0.6

‐0.1

0.4

0.9

1.4

1.9

0.00E+00 2.00E‐04 4.00E‐04 6.00E‐04 8.00E‐04 1.00E‐03 1.20E‐03 1.40E‐03

Voltage (V)

Time (s)Incident Bar Transmission Bar

‐0.12

‐0.1

‐0.08

‐0.06

‐0.04

‐0.02

0

0.02

0.00 100.00 200.00 300.00 400.00 500.00

Voltage (V)

Time (s)

Transmission Wave

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

Stre

ss (

MP

a)

Strain

0 µs 4 µs

300 µs 340 µs


Figure 10 shows the stress-strain curve comparison of a material’s raw data verses its best fit line of the strain it was tested. As viewed in the figure, the data was noisy. The noise is due to the low impendence of the material causing the signal noise to overlap with the readings collected from the transmission bar. Thus, a hollow Aluminum transmission bar was needed and used. However, the readings collected were still noisy due to how low the impedance of the material was. The materials’ responses were all plotted and the best fit trendline of a sixth order polynomial was used to create the smoothed-out data reading seen in Figure 9 and the following stress-strain graphs. Plateauing exhibited by the samples is typical for soft materials such as the foams tested in this study [4,3].

Figure 10: Raw Sample Data vs. Rounded Strain-Rate

Data

To better understand how each material responded individually to the strain-rates, Figures 11 - 13 were graphed. Figure 11 depicts the white samples’ responses to all three pressures. These samples responded with max stress values within 0.01 of each other. The values were 0.488 MPa, 0.492 MPa, and 0.500 MPa for 1800/s, 2400/s, and 2800/s respectively. These responses show a strain softening response as the strain-rates increased. Figure 12 depicts the light blue samples’ responses to the three strain-rates. It can be viewed that the samples’ responses between 1800/s and 2400/s the stress-strain curves merely extend showing the material absorbed more energy. As seen in the figure, the ultimate strength approximately the same with values of 0.504 MPa and 0.533 MPa. At 2800/s, the ultimate strength value is 0.678 MPa. The material’s entire curve shifts up on the stress axis for depicting a strain-rate sensitivity at 2800/s. In fact, light blue shows the most strain-rate sensitivity of the three materials. Between 1800/s and 2400/s there is a strain softening effect present. Meanwhile, there is strain hardening when 2800/s occurs. The strain hardening is shown on the graph by the origin, curve, and end of the 2800/s curve being shifted up. It also has an increased ultimate strength of 0.678 strain. Similarly, the maximum strain increases 30.6% between 0.49 at 1800/s to 0.64 at 2400/s. There was a 7.8% increase between the 0.64 at 2400/s and 0.69 at 2800/s. that is a total 38.4% increase from the 1800/s and 2800/s. Plateauing is shown on this graph, particularly with strain-rate 2800/s between strains 0.1 and 0.2 where the data points plateau [4, 3, 9, 10]. Figure 13 depicts the dark blue samples’ responses to all three pressures. Unlike the light blue samples, dark blue had almost a

uniform stress across all three pressures. Dark blue shows strain-rate softening. The highest strength was 0.494 MPa, 0.500 MPa, and 0.496 MPa from highest to lowest strain-rate. However, the material’s strain had slight variation. The strain values from highest to lowest strain-rate are as follows, 0.68, 0.68, and 0.51. As seen the values of the 2400/s and 2800/s strains have the same ending point at 0.68. This is a large jump of 33% from 1800/s’ 0.51.

Figure 11: White Strain-Rate Comparisons

Figure 12: Light Blue Strain-Rate Comparisons

Figure 13: Dark Blue Strain-Rate Comparisons

Figures 14 - 16 shows the comparison of all three

materials at each strain-rate. As shown in Figure 14, at 1800/s there was virtually no change in the response of the three EDPM variations. The plateauing occurred between 0.1 and 0.14 strain for all three. At this strain-rate, the densification occurred almost immediately after the plateauing from 0.19 to 0.4 strain. The

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

Stress (MPa)

StrainSample 13 Raw Dark Blue 2800/s

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

Stress (MPa)

StrainWhite 1800/s White 2400/s White 2800/s

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

Stress (MPa)

Strain (MPa)Light Blue 1800/s Light Blue 2400/s Light Blue 2800/s

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

Stress (MPa)

StrainDark Blue 1800/s Dark Blue 2400/s Dark Blue 2800/s


three materials have similar maximum strain values as well. From lowest to highest we have light blue with 0.49, white with 0.50, and then dark blue with 0.51. Not only are the three similar in compressibility, but also in strength with stress values of 0.488 MPa (W), 0.494 MPa (DB), and 0.541 MPa (LB). Figure 15 shows slight variation. White and Dark Blue have about the same maximum strengths, 0.491 MPa and 0.496 MPa. However, Light Blue has a maximum strength of 0.533 MPa. Similarly, the strain values of all three are close, yet farther apart than the 1800/s; 0.64 (LB), 0.67 (W), and 0.68 (DB). Figure 16 shows even more variation of EDPM. As viewed in the figure, the light blue’s response is higher than both dark blue and white with a value of 0.1602 strain as the beginning of the stress-strain curve. This value is about 20.5% higher than dark blue and white’s values of 0.0764 MPa and 0.0802 MPa from the beginning. As for the max stress values, from lowest to highest 0.4960 (DB), 0.5044 (W) and 0.6783 (LB). Light blue had a 1.5% higher max stress from dark blue. This can be credited to material hardening during the experiment [5]. The materials’ responses to strain also have a greater variation between them. Even though white had the highest strain with a value of 0.78. Light blue also had a lower strain value of 0.69 which is 11.5% less than white’s response. Dark blue’s strain is lowest with 0.68 which is 36.8% to light blue’s response. The numerical values of the stress and strain are in Table 1 and 2.

Figure 14: 1800/s Strain-Rate Comparisons



Table 1: Strain Values

Strain-rate White Light Blue Dark Blue Max 1800/s 0.5 0.49 0.51 0.51 2400/s 0.67 0.64 0.68 0.68 2800/s 0.78 0.69 0.68 0.78

Table 2: Stress Value

Strain-rate White Light Blue Dark Blue Max

1800/s 0.488 0.541 0.494 0.541

2400/s 0.491 0.533 0.500 0.533

2800/s 0.504 0.678 0.496 0.678

To best understand how the material varied from strain-rate and softness level Figure 17 was graphed. It depicts the material’s maximum compression strength. This shows, numerically, how much compression each material consumed. At 2800/s light blue samples responded 8.82% higher than white and 10.18% higher than dark blue. At 2400/s, light blue was 2.05% higher than dark blue and white. At 1800/s, light blue was 0.3% higher than white and 1.4% higher than dark blue. To see how the responses varied in energy absorption, Figure 18 was graphed. As viewed in the figure, the light blue samples at 2800/s is the highest with a value of 2.548 KJ/kg. Light blue also has the lowest energy absorption overall at a strain-rate of 1800/s with a specific energy value of 1.231 KJ/kg. Because it is both the lowest and highest, this material has the most strain-rate sensitivity.

Figure 17: Compressive Strengths Bar Graph

0

0.1

0.2

0.3

0.4

0.5

0.6

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

Stress (MPa)

Strain

White 1800/s Light Blue 1800/s Dark Blue 1800/s

0

0.1

0.2

0.3

0.4

0.5

0.6

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

Stress (M|Pa)

Strain


0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0 0.2 0.4 0.6 0.8 1

Stress (MPa)

Strain


0.5336 0.5476 0.5446 0.5336 0.5541 0.5336 0.5358 0.6376 0.5494

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

Stress (MPa)

240 280


Figure 17: Energy Absorption Comparisons

CONCLUSION In this study, the dynamic responses of three EDPM foams

of different softness levels (15.5, 20.0 and 25.0 Shore O) were tested in a SHPB at 1800/s, 2400/s, and 2800/s. The light blue material had both the highest and lowest impact absorption, proving it has the most strain-rate sensitivity out of the materials tested in this study. Light blue also is the only one that shows strain-rate hardening [5]. The dark blue samples had the least amount of strain-rate sensitivity. Dark blue was the most compressible at both 1800/s and 2400/s with values of 0.51 and 0.68. However, white is the most compressible at 2800/s with a strain value of 0.78.

This study provided the knowledge that using a non-contact method can be used to obtain the dynamic stress-strain response of non-rigid foams. It also highlights the need of more testing to be done, specifically on the light blue samples, for further understanding of how EPDM foams. Future works also include rubber-foam composites to see how the absorption is affected in a sandwich composite situation.

ACKNOWLEDGEMENTS The authors would like to thank McMaster-Carr for

providing the sample materials used in this research. Also connected to the samples, we would like to thank Mr. Paul Matthew Lowe for sample cutting and preparation. Similarly, thank you to the Mechanical Engineering Department of the University of Mississippi for allowing use of the laboratory facility and funding of this research. We would also like to thank Dr. Rajendran (Raj) for supporting this research and Dr. P. Raju Mantena for allowing use of the Split Hopkinson Pressure Bar.

REFERENCES [1] W. Chen, B. Z. (1999). A Split Hopkinson Bar Technique for Low-impedance Materials. Experimental Mechanics, 39(2), 81 - 85. [2] M. C. Saha, H. M. (2005). Effect of density, microstructure, and strain rate on compression behavior of polymeric foams. Materials Science and Engineering , 328-336. [3] Simon Ouellet, D. C. (2006). Compressive Response of Polymeric Foams Under Quasi-Static, Medium and High Strain rate Conditions. Science Direct, 731-743.

[4] J. W. Tedesco, C. A. (1993). Strain rate Effects on the Compressive Strength of Shock-Mitigating Foams. Journal of Sound and Vibration, 376 - 384. [5] Daniel Whisler, H. K. (2015). Experimental and simulated high strain dynamic loading of polyurethane foam. Polymer Testing 41, 219-230. [6] William D. Callister, J. (2007). Materials Science and Engineering an Introduction. New York: John Wiley & Sons Inc. [7] D.j. Frew, M. J. (2002). Pulse Shaping Techniques for Testing Brittle Materials with a Split Hopkinson Pressure Bar. Experimental Mechanics, 93-106. [8] Siviour, C. a. (2016). High Strain-rate Mechanics of Polymers: a Review. Journal of Dynamic Behavior of Materials. [9] Remy Bouiz, P. V.-L. (2009). Polypropylene foam behaviour under dynamic loadings: Strain-rate, density and microstructure effects. International Journal of Impact Engineering, 329-342. [10] Kim, D. W. (2015). Experimental and simulated high strain dynamic loading of polyurethane foam. Polymer Testing, 219-230.

1.408 1.928 1.328 1.231 1.694 2.548 1.471 1.999 2.476

0

0.5

1

1.5

2

2.5

3

Spec

ific E

ner

gy (

KJ/

Kg)

DB 1800/s DB 2400/s DB 2800/s LB 1800/s LB 2400/s LB 2800/s W 1800/s W 2400/s W 2800/s





DYNAMIC RESPONSE OF POROUS METALLIC AND POLYMERIC FOAMS FOR ENHANCING ENERGY ABSORPTION UNDER HIGH STRAIN-RATES

Suman Babu Ukyam University of Mississippi

University, Mississippi USA

Damian Stoddard University of Mississippi


P. Raju Mantena University of Mississippi


ABSTRACT Light weight sandwich composite panels used in aerospace and secure building construction are likely to be subjected to high energy

ballistic and blast loads. In this paper, the dynamic response of porous metallic and polymeric foams evaluated at high strain-rates, is reported. Three different density variants (75, 110 and 205 Kg/m3) of closed-cell rigid ROHACELL® HERO polymeric foam; and two different densities (4-6% and 10-14%) of open-cell porous aluminum Duocel® foam, with three variations in pore size (10, 20 and 40 PPI) were characterized at the same relative density. Dynamic compression tests were performed on these low-impedance materials with a hollow transmission bar in a Split-Hopkinson Pressure Bar (SHPB) set-up. Strain-gage output from the hollow transmission bar was also validated with a non-contact digital image correlation technique (DIC).

For the polymeric foams it was observed that increasing density exhibited higher strength and stiffness with plateau stress level increasing proportionately but with a reduction in strain range. High density foams were subjected to lower strain rates compared to low density foams at similar loading conditions. No change in dynamic response of the metallic foams was observed when comparing the three different porosities with same relative density at similar strain rates. Based on the results of these investigations, functionally grading the core material in a sandwich composite with varying density foams for improving energy absorption while maintaining structural integrity, is ongoing.

KEY WORDS: SHPB, ROHACELL® HERO, Duocel®, Digital Image Correlation.

This work was submitted to UAB ECTC 2019 for presentation only





COMPARATIVE ANALYSIS AND DYNAMIC RESPONSE OF GAROLITES UNDER TEMPERATURE SPECTRUM USING LOW VELOCITY IMPACT TEST

Birendra Chaudhary University of Mississippi Oxford, Mississippi USA

Suman Babu Ukyam University of Mississippi Oxford, Mississippi USA

Damian Stoddard University of Mississippi Oxford, Mississippi USA

ABSTRACT The purpose of this study was to examine the dynamic

response of three different grades of garolites under temperature spectrum. High Temperature G-11 Sheet (HTG), Impact Resistance Garolite E-glass (HIG) and Economical Garolite (EG) were tested using an Instron Dynatup 8250 impact tester. Three specimens were used for all three composites and were tested at 6 different temperatures, -10oC, 25oC, 50oC, 100oC, 150oC and 200oC, using the Low Velocity Impact Machine with 20 Kip punch shear load cell. The results showed that HIG had the highest resistance to punch shear impact. It resisted the highest amount of impact followed by HTG and EG at every temperature tested. The total energy absorbed by HIG was roughly 12 times EG and roughly thrice as much as HTG. The damage propagation energy of HIG was roughly 14 and 3 times those of EG and HTG. Over the temperature spectrum, it was observed that the energy absorption of HIG until peak load was around 11 times and 4 times the energy absorption of EG and HTG respectively. The maximum impact load for HIG was respectively around 5 times and twice those of EG and HTG respectively. Similarly, the maximum impact absorbed by each Garolite decreased with increases in temperature. Also, the failure zone decreased with increases in temperature.

KEY WORDS: High Temperature Garolite, High Impact Garolite, Economical Garolite, Low Velocity Impact Test, Impact Absorption, Punch Shear

INTRODUCTION Composite materials provide great benefits because of their

high strength-to-weight ratio, compressive strength, corrosion resistance, fatigue resistance, and non-magnetic properties [1]. However, they are vulnerable to damage from low-velocity impact (LVI). Impact may cause any combination of damage modes including fiber crushing, delamination, through thickness

shear fracture, and perforation [1]. When composite materials were subjected to mechanical loading and exposed to severe environmental conditions, the natural fiber reinforced composites seemed reasonably strong and had the potential to be used as materials for strong components such as automotive, building materials, shipping etc., although they had some limitations when compared to reinforced glass, such as high moisture absorption and lower strength [2,3].

Garolite is a woven Fiberglass-epoxy laminate material. It is created by stacking layers of glass cloth, soaking in epoxy resin, and compressing the resulting material under heat and pressure until the epoxy cures. Because both Micarta and carbon fiber laminates are resin based laminates, they are similar to Garolite except for the base material, which is glass cloth [4]. This material has dimensional stability, high strength over temperature, very negligible moisture absorption and high level of electrical insulation and chemical resistance, so it is used in several aerospace applications, circuit boards, machinery equipment etc. [4,5]. Carbon fiber composites can be replaced by garolite due to similar composition and properties at a fraction of its cost.

Recently Fei Zhou et al. reported that the strength decreased with increase in temperature of the Carbon Fiber Reinforced Polymer (CFRP) tendons due to the softening and decomposition of the resin, which weakened the bonding effect of fibers [6]. Another study by B.C.Ray on interfaces of glass and carbon fiber reinforced epoxy composites resulted in a significant weakening that often appeared at the interface during hygrothermal ageing [7]. A work by T. Gomez-del Rio on response of carbon fiber reinforced epoxy matrix (CFRP) laminates at LVI on low temperature suggested that the damage induced in those laminates increased with impact energy. It also stated that cooling the laminate before the impact had an effect on damage similar to that of increasing the impact energy [8].

Many investigators have concluded that the fiber reinforced composites are effective members for concrete members.


Pneumatic Chamber

Spring Assists

Guide Rails

Trigger Mechanism

Load cell

Tup

Clamping Fixture

However, the challenge still exists in the increasing application of those composites, such as fully understanding material properties of fiber composites at higher temperatures [9,10]. Very few studies have been done regarding the high temperature effect on the mechanical properties of the fiber composites, which is indeed needed. So, to present a better understanding the dynamic response of the fiber reinforced composites, this research used three different grades of Garolite subjected to LVI testing for further study.

In this study, an LVI machine was used to impact specimens and create a punch shear loading scenario. This method is often used in order to focus on the unique impact damage behavior in a material [11]. An LVI test is very different from a high impact velocity test or a quasi-static test. For LVI, the contact duration is long enough for the entire structure to respond to the impact, and energy is absorbed elastically and/or eventually in damage creation, whereas for high velocity, the impact event is short and the structure may have no time to respond in flexural or shear modes [12]. The quasi static test is performed at a very slow rate such that the internal equilibrium of the specimen is maintained. In this experiment, it is expected that the maximum impact energy absorbed by the garolite composites will decrease over the elevated temperatures.

METHODS AND MATERIALS All three Garolites were obtained McMaster-Carr. They

were ordered as sheets of 30 cm x 30 cm x 0.625 cm. The materials were then sized to fit into the Low Impact Velocity Machine and were milled using the Saw machine at the Machine Shop at The University of Mississippi. Samples dimensions to approximately 10 cm x 10 cm were prepared for the testing. Figure 1 shows the sample specimens for each garolite kept at room temperature (25oC).

To identify the garolites’ mechanical properties at varying temperatures, these samples were heated to 200oC and tested in the LVI machine. Similarly, to test the materials at -10oC, an industrial freezer at the Center of Manufacturing Excellence, was used. The freezer was kept at a constant temperature of -25oC. An ice bath was prepared to transport the samples from the freezer to the LVI machine to keep the samples from gaining too much ambient heat from the surroundings. Samples were sealed inside a plastic bag while in the ice bath.

Figure 1. Sample Specimens for High Temperature G-11 (left), Economical Garolite (middle) and Impact

Resistance Garolite (right)

Since the temperature of the heated samples and the cooled

sample were different than the ambient temperature, heat loss (for the samples at higher temperature) and heat gain (for the samples at lower temperature) would occur. Due to this phenomenon, the samples at higher temperature were heated to higher temperature than required temperature to counteract the heat loss. For this, an estimation of 40 seconds to place the samples under the clamping fixture and test was used.

It was assumed that the internal resistance of the body (conduction) was negligible in comparison with the external resistance (convection). The Lumped Heat Capacity Formula was used to calculate the temperature the samples would need to obtain inside the oven before testing. Also, the time the samples would require to maintain a uniform temperature both inside and outside was calculated [13]. An infrared thermometer was used to measure the temperature of the samples.

EXPERIMENTAL SETUP All the impact tests for 3 different grades of garolites at 6

different temperatures were conducted on an Instron Dynatup 8250, the LVI Machine with the pneumatic rebound brake system at the Structure and Dynamics Lab at the University of Mississippi. A 20 Kips load cell was used for puncturing through the samples with hemispherical tip of roughly 12.7 mm. The Pneumatic assist force was kept at 80 Psi throughout the entire testing for consistency. Specimens were impacted with the load mass of 35 Kg, and the impact velocity and impact energy being roughly 5.7 m/s and 565 J respectively. Similarly, the clamping force was kept at 80 Psi to minimize the movement of samples during impact. Figure 2 shows the experimental setup of the LVI Machine.

Figure 2 Low Velocity Impact Machine


A Shimadzu HPV-2 High-Speed Video Camera with a fixed resolution of 312 x 260 pixels and recording speed of 32,000 frames per second was used to capture the impact. The illumination was provided by a GS Vitech MultiLED QT system to capture clear images and provide enough lighting during impact due to very low exposure time. Figure 3 shows a clearly focused image of HTG captured by Shimadzu HPV-2, clamped in between the clamping configuration just before the impact and illuminated by GS Vitech MultiLED QT system.

Figure. 3 Imaging of High Temperature Garolite with the

Use of Lighting

A personal computer-based data acquisition system, supplied by Dynatup, was triggered by a photo diode velocity detector just prior to impacting the specimen and was used to collect data from the load cell tup. The rebound brake was also triggered by the velocity detector and engaged after the initial impact to prevent multiple impacts on the test specimens [14]. The specimen was fixed in the steel clamp. It was indented with the indenter tup of radius 12.7 mm. During testing, a linear variable displacement transducer mounted under the specimen recorded the displacement of the center of the indenter. For each test, the load versus indenter displacement data was collected via a digital data acquisition system. The free-falling impactor was allowed to fall along two smooth guided columns upon release, and the total displacement of impactor and top skin deflection were recorded as a function of time with a data acquisition system [15].

The standard used to conduct the testing was (American Society of Testing and Materials) ASTM D3763-10, a standard method for high speed puncture properties of plastics using load and displacement sensors. According to this standard, the impact energy was kept over thrice the required energy to fully puncture the specimen to keep the velocity slowdown within 20% [16].

RESULTS The impact tests were performed using the LVI machine for

three different Garolites, HTG, HIG and EG. Each material was tested at 6 different temperatures, and 3 specimens of each garolite were tested at each temperature.

In general, the maximum load a specimen can withstand decreased with increase in temperature for all three Garolites

with lowest being the load at 200oC for each Garolite, which can be seen in Figure 4, Figure 5 and Figure 6 respectively for HTG, EG and HIG. However, the maximum load increased from -10oC to 25oC (room temperature). This could be because the materials, when manufactured, are aimed to work best at the normal room temperature, and as the material goes to high temperature, they degrade causing the loss of impact load resistance. The max load absorbed by the specimens under the different temperature spectrum is listed in Table 1. Similarly, a column chart to compare the max impact load at each temperature for all the garolites is shown in Figure 7.

Figure 4. Load vs Deflection Curve of High Temperature

Garolite

Figure 5. Load vs Deflection Curve of Economical Garolite

0

5000

10000

15000

20000

25000

0 5 10 15 20 25 30 35

Load

(N)

Deflection (mm)

HTG ‐10˚ C

HTG 25˚ C

HTG 50˚ C

HTG 100˚ C

HTG 150˚ C

HTG 200˚ C

0

5000

10000

15000

20000

25000

0 5 10 15 20 25 30 35

Load

(N)

Deflection (mm)

EG ‐10˚ C

EG 25˚ C

EG 50˚ C

EG 100˚ C

EG 150˚ C

EG 200˚ C


Figure 6. Load vs Deflection Curve of High Impact Garolite

Table 1: Max Impact Load in Joules for all three grades of

garolite at various temperatures

‐10 oC 25 oC 50 oC 100 oC 150 oC 200 oC

HTG 12328.07 16299.83 12727.33 13834.27 10572.9

0 8290.7

7

HIG 21423.04 21040.40 21682.17 21081.90 20120.5

0 19694.

73

EG 4253.45 5255.23 4798.57 4584.97 3664.63 3377.1

0

Figure 7. Max load at different temperature

The load withstood by EG and HTG was maximum at room

temperature. A downward trajectory can be seen after the room temperature as the temperature increases. This could be because material degradation of the resin caused weak bonds and softening at increased temperature. However, a large drop in

impact load for HTG from 25oC to 50oC was seen. This could be due to the inconsistency in the samples, which is fairly typical among the composites. A fairly consistent maximum load was seen for HIG up-to 50oC, followed by a downward trajectory. The impact load seems to be higher at -10oC than at 25oC but is within the statistical spread and needs more investigation.

A side by side comparison of each Garolite for Load vs Deflection at a fixed temperature showed that HIG absorbed a higher amount of impact than the other composites, with EG Garolite being the weakest. A similar trend was seen at each temperature tested, which confirms that HIG was indeed the strongest among the test samples. HIG is a woven material which has been shown to resist high amounts of impact, providing better energy absorption. The load vs deflection curves for different garolites at -10oC, 25oC, 50oC, 100oC, 150oC and 200oC are shown in Figures 8, 9, 10, 11, 12 and 13 respectively.

Figure 8. Side by side load vs deflection comparison of

garolites at -10oC

Figure 9. Side by side load vs deflection comparison of garolites at 25oC

0

5000

10000

15000

20000

25000

0 5 10 15 20 25 30 35

Load

(N)

Deflection (mm)

HIG ‐10˚ C

HIG 25˚C

HIG 50˚C

HIG 100˚C

HIG 150˚C

HIG 200˚C

0

5000

10000

15000

20000

25000

Load

(N)

Temperature (oC)HTG HIG EG

‐10 5025 150100 200

0

5000

10000

15000

20000

25000

0 5 10 15 20 25 30 35

Load

(N)

Deflection (mm)

HIG

HTG

EG

0

5000

10000

15000

20000

25000

0 5 10 15 20 25 30 35

Load

(N)

Deflection (mm)

HIG

HTG

EG



Figure 11. Side by side load vs deflection comparison of

garolites at 100oC



It can be seen that the spread of the load vs deflection graph

for all three garolites at all temperatures were significantly different. This is due to the failure pattern of the garolites. Huge deflection was seen on HIG samples before shear punch through. Due to this failure pattern, it absorbed the most energy and is ductile. Some resistance was seen on HTG samples before punch through, because cracks caused more energy absorption than EG but less than HIG. Also, no visible cracks or deflection were seen on EG samples, and the shear punch was seen as soon as the load cell hit the sample, due to which the sample absorbed less energy, making it more brittle than HTG and HIG.

Similarly, the energy to max load and damage propagation energy of each Garolite until failure at different temperature are listed in Tables 2 and 3 respectively with a respective column chart in Figures 14 and 15.

Table 2. Energy to Max Load in Joules for all three grades

of garolite at various temperatures

‐10 oC 25 oC 50 oC 100 oC 150 oC 200 oC

HTG 42.78 50.04 52.96 41.52 27.17 38.08

HIG 159.68 150.09 160.92 171.67 175.48 147.31

EG 11.07 11.07 14.66 19.36 22.18 11.71

Table 3. Damage Propagation Energy in Joules for all three grades of garolite at various temperatures

‐10 oC 25 oC 50 oC 100 oC 150 oC 200 oC

HTG 51.51 90.73 58.64 88.32 85.59 55.67

HIG 188.43 200.55 183.14 193.69 174.17 176.73

EG 12.18 19.76 21.36 15.34 6.43 11.67

0

5000

10000

15000

20000

25000

0 10 20 30 40

Load

(N)

Deflection (mm)

HIG

HTG

EG

0

5000

10000

15000

20000

25000

0 5 10 15 20 25 30 35

Load

(N)

Deflection (mm)

HIG

HTG

EG

0

5000

10000

15000

20000

25000

0 5 10 15 20 25 30 35

Load

(N)

Deflection (mm)

HIG

HTG

EG

0

5000

10000

15000

20000

25000

0 5 10 15 20 25 30 35

Load

(N)

Deflection (mm)

HIG

HTG

EG


Figure 14. Energy to Max Load

Figure 15. Damage Propagation Energy

Looking at the Energy to max load chart, it can be seen that

there was a general increasing trend for HIG and EG up to 150oC. The reading at -10oC does not follow the trend and hence needs further investigation. The decrease at 200oC could be due to the brittle transition of the material or to degradation of the resin. The large error bar for HIG at 150oC could be due to inconsistencies in the samples and needs more testing. Similarly, an upward trend was seen up to 50oC and a downward trend up to 150oC for HTG. This could be due to the fact that HTG’s can only work up-to a certain high temperature and may not react well to extreme temperatures above 150oC. The sudden increase at 200oC could be due to the inconsistency in the samples and requires further investigation. It can also be seen that HIG propagates the highest amount of energy among all these garolites, even at high temperatures. HIG would provide better impact resistance in high temperature applications. The damage propagation is more dependent on the damage mechanism than on temperature and hence none of the garolites showed a consistent pattern.

Table 4 and column chart in Figure 16 represents the total energy absorption in Joules by the Garolites at different temperatures.

Table 4. Total Energy Absorption in Joules for all three

grades of garolite at various temperatures

‐10 oC 25 oC 50 oC 100 oC 150 oC 200 oC

HTG 94.31 140.78 111.60 129.84 112.76 93.75

HIG 348.11 350.64 344.06 365.36 349.64 324.04

EG 23.24 30.82 36.03 34.70 28.61 23.39

Figure 16. Total Energy Absorption in Joules

It can be seen that HIG absorbed the highest amount of

energy while EG absorbed the least. This could be due to the failure pattern of these garolites and the configuration of the layers. HIG is a woven material, and deflection was seen before delamination during the test which can be seen in Figure 17. This helps in the absorption of a huge amount of energy. Similarly, crack propagation was observed during shear puncture for HTG, and a punch through for EG, which can be seen in Figures 18 and 19 respectively. Due to the shear punch through, EG does not absorb much energy, while HTG absorbs some energy during damage propagation through cracks. The total energy absorbed by HIG was fairly consistent throughout all the temperatures, with a slightly decreasing trend after 150oC. Similarly, HTG absorbed the highest energy at 25oC and showed a decreasing trend with the increase in temperature except at 50oC which showed a large dip in the values. This could be due to the failure pattern or to inconsistency in the samples. As for EG, the total absorption was fairly consistent and did not change much due to its nature. The total energy it absorbed was comparatively lower than the others and hence the change due to increase in temperature was minimum. However, an increasing trend up-to 50oC, followed by a decreasing trend with the increase in temperature was seen.

0

50

100

150

200

250

300

350

400

Energy (J)

Temperature (oC)

HTG HIG EG

‐10 5025 150100 200

0

50

100

150

200

250

300

350

400

Energy (J)

Temperature (oC)

HTG HIG EG

‐10 5025 150100 200

0

50

100

150

200

250

300

350

400

Energy (J)

Temperature (oC)

HTG HIG EG

‐10 5025 150100 200


Figure 17. Damage Propagation on Impact Resistance Garolite at 50oC right after impact

Figure 18. Damage Propagation on High Temperature Garolite at 50oC right after impact

Figure 19. Damage Propagation on Economical Garolite at 50oC right after impact

It can also be observed from Figure 20, which shows the failure zone of the garolites after impact, that the failure zone relatively decreased with the increase in temperature. Further investigation would be required to understand the underlying mechanisms causing the phenomenon.

Figure 20. Failure zone analysis of High Impact (top), Economical (middle) and High Temperature (bottom)

Garolite at -10oC, 25oC, 50oC,100oC,150oC and 200oC from left to right respectively

CONCLUSION

In this study, the dynamic response of three different garolites, High Temperature G-11 Sheet, Impact Resistance Garolite E-glass and Economical Garolite Sheet were tested at 6 different temperatures, -10oC, 25oC, 50oC, 100oC, 150oC and 200oC. HIG showed the highest impact absorption, whereas EG showed the lowest strength on impact. The total energy absorbed by HIG was roughly 12 times that of EG and roughly thrice as much as HTG. Similarly, the damage propagation energy of HIG was roughly 14 and 3 times those of EG and HTG. In general, for each garolite, as the temperature increased, the maximum load decreased starting at room temperature (25oC), except for HIG which started decreasing after 50oC. The maximum load increased from -12oC to 25oC, as the materials tend to work best at the room temperature. Overall, the energy absorbed by HIG until max loading condition was around 11 times and 4 times the energy absorbed until maximum loading condition by EG and HTG respectively. The maximum impact load for HIG was respectively around 5 times and twice EG and HTG.

This study provided some important information regarding the impact response of garolites when subjected to significantly higher temperatures. There are several future aspects of this experiment. Some inconsistencies were seen throughout the experiment which can be explored and refined in future works. Similarly, the dynamic response of these specimens can be tested using high impact velocity test and quasi static test. An investigation to fully understand the mechanism that causes the failure zone to decrease with the increase in temperature in garolite can be done in future works. These materials can be used for military applications such as making barricades, and they weather out over time. So, studies regarding the low velocity and


high impact test for weathered garolites can be done in future to better understand the change in dynamic response of these materials after they are weathered.

ACKNOWLEDGEMENTS This research was partially funded by the US Army Corps

of Engineers- Engineer Research and Development Center (ERDC)Vicksburg, under primary contract #W912HZ18C0025, program manager Dr. Robert Moser, Senior Research Civil Engineer - Materials US Army ERDC GSL. The authors would like to thank McMaster-Carr for providing three different grades of garolites used in this research. Similarly, we would also like to thank Mr. Paul Matthew Lowe for sample preparation and The Department of Mechanical Engineering, The University of Mississippi for allowing the use of the laboratory and providing partial funding for the research. We would also like to thank Dr. Rajendran (Raj) for supporting this research and Dr. P. Raju Mantena for allowing us to use the LVI Machine.

REFERENCES [1] Lawrence, B. D., & Emerson, R. P. (2012). A Comparison of Low-Velocity Impact and Quasi-Static Indentation. doi:10.21236/ada579696 [2] Ghani, M., Salleh, Z., Hyie, K. M., Berhan, M., Taib, Y., & Bakri, M. (2012). Mechanical Properties of Kenaf/Fiberglass Polyester Hybrid Composite. Procedia Engineering,41, 1654-1659. doi:10.1016/j.proeng.2012.07.364 [3] Hazizan MdAkil, Leong Wei Cheng, ZA. Mohd Ishak, A. Abu Bakar, M. A. Abd Rahman, 2009. Water Absorption Study on Pultruded Jute Fiber Reinforce Unsaturated Polyester Composite, Composites Sciences and Technology, 691942-1948 [4] (n.d.). Retrieved from http://www.matweb.com/search/datasheetText.aspx?bassnum=PGLAM04 [5] Song, Z., Wang, Z., Ma, H., & Xuan, H. (2014). Mechanical behavior and failure mode of woven carbon/epoxy laminate composites under dynamic compressive loading. Composites Part B: Engineering,60, 531-536. doi:10.1016/j.compositesb.2013.12.060 [6] Zhou, F., Zhang, J., Song, S., Yang, D., & Wang, C. (2019). Effect of Temperature on Material Properties of Carbon Fiber Reinforced Polymer (CFRP) Tendons: Experiments and Model Assessment. Materials,12(7), 1025. doi:10.3390/ma12071025 [7] Sharma, N., Kumar, M. S., & Ray, B. (2008). Study of the Effect of Hygrothermal Ageing on Glass/Epoxy Micro-Composites by FTIR-Imaging and Alternating DSC Techniques. Journal of Reinforced Plastics and Composites,27(15), 1625-1634. doi:10.1177/0731684407086318 [8] Rı́o, T. G., Zaera, R., Barbero, E., & Navarro, C. (2005). Damage in CFRPs due to low velocity impact at low temperature. Composites Part B: Engineering,36(1), 41-50. doi:10.1016/j.compositesb.2004.04.003 [9] Si-Larbi, A., Agbossou, A., Ferrier, E., & Michel, L. (2012). Strengthening RC beams with composite fiber cement plate reinforced by prestressed FRP rods: Experimental and numerical

analysis. Composite Structures,94(3), 830-838. doi:10.1016/j.compstruct.2011.10.001 [10] Zhang, S., Yu, T., & Chen, G. (2017). Reinforced concrete beams strengthened in flexure with near-surface mounted (NSM) CFRP strips: Current status and research needs. Composites Part B: Engineering,131, 30-42. doi:10.1016/j.compositesb.2017.07.072 [11] Paradiso, A. E., & Lamberson, L. (2018). Failure Behavior of Woven Fiberglass Composites Under Combined Environmental and Mechanical Compressive Loading. Philadelphia, PA: Drexel University. [12] Donadon, M., Iannucci, L., Falzon, B., Hodgkinson, J., & Almeida, S. D. (2008). A progressive failure model for composite laminates subjected to low velocity impact damage. Computers & Structures,86(11-12), 1232-1252. doi:10.1016/j.compstruc.2007.11.004 [13] Lienhard, J. H., Lienhard, J. H. (2003). A heat transfer textbook. 3rd ed. Cambridge, Mass.: Phlogiston Press. [14] Schoeppner, G., & Abrate, S. (2000). Delamination threshold loads for low velocity impact on composite laminates. Composites Part A: Applied Science and Manufacturing,31(9), 903-915. doi:10.1016/s1359-835x(00)00061-0 [15] Zhu, S., & Chai, G. B. (2013). Damage and failure mode maps of composite sandwich panel subjected to quasi-static indentation and low velocity impact. Composite Structures,101, 204-214. doi:10.1016/j.compstruct.2013.02.010 [16] ASTM D3763-10,2010, “Standard Test Method for High Speed Puncture Properties of Plastics Using Load and Displacement Sensor,” ASTM International, West Conshohocken, PA, 2003, DOI: D20.10/D3763-10, www.astm.org.

ectc section 2 cover...composite beams”, i negru, g r gillich, z i praisach, m tufoi, n gillich....

Documents