The energy-absorbing characteristics of polymer foams reinforced with bamboo tubes

Abstract

This paper investigates for the first time, the energy-absorbing characteristics of a range of lightweight bamboo-reinforced foam structures. Initial attention focuses on characterizing the energy-absorbing characteristics of the individual bamboo tubes and assessing the influence of tube geometry on energy absorption. Here, it has been shown that the specific energy absorption of the bamboo tubes decreases with increasing tube length as a result of axial splitting associated with barreling of the samples under compressive loading. The influence of the tube inner diameter/thickness (D/t) ratio was also investigated using a number of tube sizes, where it was shown that there is a small increase in specific energy absorption with decreasing D/t. The tubes were then embedded in crosslinked PVC foams in order to investigate the influence of varying degree of external support applied to the reinforcement on the failure modes in the tubes as well as the measured specific energy absorption values.

Finally, alternative techniques for enhancing the energy-absorbing capacity of the tubes were investigated. Here, tubes of different length were wrapped circumferentially in epoxy-impregnated kenaf fibers to enhance their resistance to axial spitting. It was shown that reinforcing the tubes in this manner can significantly enhance the ability of the tubes to absorb energy under conditions of axial quasi-static crushing.

Keywords

Foam fracture bamboo impact sandwich

Introduction

With the increasing need to develop lightweight, energy-absorbing structures for use in high-performance engineering design, researchers have, in recent years, been investigating the potential offered by simple composite structures, such as tubes based on carbon, Kevlar and glass fiber reinforced epoxy resin [1 –4]. Previous work has shown that through careful design and the application of appropriate loading conditions, composite materials can offer extremely high values of energy absorption under both quasi-static and dynamic loading conditions [5,6]. In general, progressive failure in axially loaded composite tubes can be achieved through the introduction of an appropriate triggering mechanism, typically a chamfered profile at one end of the tube [7 –9]. Previous work has highlighted a multitude of failure mechanisms in composite tubes undergoing progressive crushing, including transverse shearing, local buckling and progressive folding, longitudinal and interlaminar cracking, fiber fracture as well as splaying and lamina bending [6,10 –15]. The relative amount of each of these energy-absorbing mechanisms occurring during crushing depends on a number of factors including the properties of the fiber, matrix and associated interface as well as geometric effects and loading conditions [3,6].

Typically, the energy-absorbing capacity of a composite tube is assessed by determining its specific energy absorption (SEA) value, based on the total energy absorbed in crushing the tube per unit mass of material. Typical values for SEA following tests on epoxy-based composites range from 39 kJ/kg for a pultruded glass fiber reinforced polyester to 110 kJ/kg for a carbon fiber reinforced epoxy system [6], with even higher values being recorded following tests on thermoplastic-matrix materials [16]. Following tests on a range of tube diameters, Farley [17] showed that the measured values of SEA for both carbon and Kevlar-based tubes are strongly dependent on the ratio of the inner diameter of the tube, D, to its thickness, t. He showed that decreasing the value of D/t resulted in a significant increase in specific energy absorption and attributed this effect to an increase in interlaminar cracking in the crush region. Recent work has investigated the possibility of profiting from this geometrical dependency by embedding small diameter composite tubes in a low-density polymer foam [18]. Here, carbon fiber tubes, manufactured using a roll-wrapping procedure, were embedded in a range of polymer foams and crushed at low and high rates of loading. A subsequent comparison of the data with a broad range of lightweight core materials indicated that tube-foam structures offer significant potential for use in lightweight energy-absorbing applications [18]. Ji and co-workers investigated the energy-absorbing characteristics of hybrid cores consisting of metallic multitubes and a polymer matrix [19,20].

In recent years, there has been an increasing drive to develop environmentally friendly materials and lightweight structures for use in a wide range of engineering applications. Research has shown that when compared in terms of their specific properties (i.e. when normalized by their respective density) natural fiber-based composites offer a range of mechanical properties that are comparable to those of more traditional composite materials [21 –23]. Much of this work focuses on evaluating the in-plane properties of natural fiber composites, particularly at low rates of strain. Although a number of studies have investigated their impact behavior, there have been few attempts to develop natural fiber composite structures for use in specific energy-absorbing applications. Meredith et al. [24] manufactured and tested a range of conical structures based on jute, hemp and flax fibers in an epoxy matrix. The authors showed that there was a strong correlation between the measured value of SEA and fiber volume fraction, with values for the hemp fiber composite being similar to those measured on a more conventional carbon fiber composite. The authors observed significant scatter in some samples, due primarily to variations in both fiber strength and fiber volume fraction. Mahdi et al. [25] tested a range of cotton fiber reinforced epoxy tubes at quasi-static rates of strain and measured values of SEA that ranged between 10 and 13 kJ/kg. The authors concluded that the measured load-displacement traces were strongly dependent on the geometry of the test specimen as well as the fiber orientation. Finally, Ataollahi et al. [26] manufactured and tested square tubes based on silk fiber reinforced epoxy resins and studied the effect of tube length on SEA. The authors reported relatively low values of SEA, typically in the range of 4 to 5 kJ/kg.

The work presented here aims to combine the benefits associated with the use of environmentally friendly materials with the attractive energy-absorbing properties of relatively small tubular structures. Here, bamboo tubes are embedded in a range of crosslinked PVC foams to develop a novel range of energy-absorbing core materials. In particular, the paper focuses on understanding the influence of core properties on energy absorption as well as identifying techniques for further enhancing the energy absorption of the bamboo tubes.

Experimental procedure

The initial part of this investigation focused on assessing the energy-absorbing characteristics of individual unchamfered bamboo tubes. Here, tubes with nominal external diameters between 10 and 24 mm and values of inner diameter to thickness ratio between 3.3 and 6.0 were cut from bamboo canes using a band saw. In order to investigate geometrical effects in the energy-absorption characteristics of the bamboo, tubes with lengths of 10, 20, 30 and 40 mm were prepared. Geometrical effects were further investigated by examining the influence of varying the tube D/t ratio on SEA. This was assessed using 20-mm-long tubes with D/t values between 3.2 and 6.0.

Compression tests were undertaken on an Instron 5969 universal test machine. Initially, the crosshead displacement rate was set at 1 mm/minute although this was increased to 5 mm/minute during the latter stages of crush. Where appropriate (i.e. if the tubes had not failed prematurely), the tests were interrupted when the nominal strain (based on the crosshead displacement and the initial length of the tube) reached 75%. The specific energy absorption (SEA) of each bamboo tube was calculated from the energy under the load-displacement trace, U, and the mass of the crushed tube, m [6].

i . e . SEA = U / m

Following testing, the samples were photographed in order to elucidate the prevailing failure mechanisms.

The energy-absorbing capacity of bamboo reinforced foam was investigated by embedding 20-mm-long tubes in four crosslinked PVC foams of similar thickness. In principle, it should be possible to form the foam around the bamboo tubes in order to give more intimate contact between the two types of constituent materials. This was not undertaken here since it was believed that it would be easier to vary the density of the foam by using foam sheets. During these tests, the nominal external diameter of the tubes was 24 mm and the average D/t value was 6. The nominal densities of the foams ranged from 40 kg/m³ to 200 kg/m³, thus enabling the influence of the stiffness of the support on the energy-absorbing behavior of the tubes to be investigated. Details of mechanical properties of the foams are given in Table 1 [27]. The planar dimensions of the foam samples were 75 × 75 mm and a hole of similar diameter to that of the bamboo was introduced into the center for insertion of the tubes, Figure 1. This procedure resulted in a tolerance fit between the foam and the bamboo. As before, all of the tubes were unchamered (to minimize sample preparation time). Tests were also conducted on foam blocks having a similar mass to those used above. These tests were undertaken to determine and then remove the contribution of the supporting foam in the energy-absorbing characteristics of the tube-reinforced cores. Compression tests on the foam-based samples were undertaken at the crosshead displacement rate employed for testing the plain bamboo samples.

Table 1.

Nominal densities and mechanical properties of the crosslinked PVC foams investigated in this study.²⁷

Foam designation	Nominal foam density (kg/m³)	Compression modulus (MPa)	Compression strength (MPa)	Shear modulus (MPa)
C70/40	40	41	0.45	13
C70/100	100	130	2.0	40
C70/130	130	170	3.0	54
C70/200	200	280	5.2	75

During the final stages of this investigation, techniques for enhancing the energy-absorbing characteristics of the bamboo tubes were evaluated. Here, the effect of reinforcing the tubes in the circumferential direction was assessed by manually wrapping kenaf fibers around a number of tubes. The modulus of the kenaf fibres typically lies between 23 and 27 GPa and their strength values between 427 and 519 MPa. It is worth noting that the corresponding values for bamboo are 18–55 GPa for modulus and 391–713 MPa for tensile strength [28]. Here, the outer surfaces of the bamboo rods were lightly abraded and then coated with a layer of epoxy resin. The thickness of the kenaf fibre reinforced epoxy layer was approximately 3.3 mm with a variation of +/- 0.3 mm. The kenaf fibers were then dipped in epoxy resin and wrapped around the outer surface of the tubes whilst continuously applying additional resin. The weight fraction of kenaf fibers in this outer composite layer was estimated to be 40%.

Results and discussion

Compression tests on the bamboo tubes

Initial attention focused on establishing the overall level of scatter exhibited by the bamboo tubes. Here, 10 tubes with a nominal length of 20 mm (average D/t = 6) were crushed under identical conditions. Figure 2 summarizes the maximum force values as well as the resulting values of SEA for these tubes. An examination of the figure indicates that the measured values are reasonably consistent given that these are naturally occurring materials. For example, the maximum force values vary between approximately 20 and 25 kN and the measured values of SEA vary between 17.5 and 20.8 kJ/kg. These values of SEA are clearly lower than those measured on carbon fiber reinforced epoxy composites, for which values typically lie between 50 and 90 kJ/kg. They do, however, compare favorably with values measured following tests on a pultruded glass fiber/polyester composite (20 kJ/kg) and a Kevlar fiber/epoxy composite (14/kJ/kg) [29]. These observations are nevertheless encouraging and suggest that these materials offer some potential for the design of environmentally friendly energy-absorbing structures.

The next part of this study focused on establishing the influence of tube geometry on the energy-absorbing capacity of the individual bamboo tubes. Initially, the influence of the tube inner diameter/thickness ratio was assessed using five test configurations. Figure 3 shows the variation of SEA with D/t where greater levels of scatter are apparent in the data area than in Figure 2. In spite of this increased scatter, the figure does, however, suggest an increase in energy absorption with decreasing D/t. For example, the value of SEA for the smallest value of D/t is approximately 33% greater than that of its largest counterpart. These findings concur with observations following tests on carbon fiber reinforced epoxy tubes, trends that were associated with a change in failure mode as the specimen geometry was changed. Following testing, the bamboo tubes were examined in detail in order to elucidate the prevailing failure modes. Here, all of the samples failed predominantly as a result of longitudinal splitting, with there being little evidence of a change in failure mode with varying geometry. The failure processes occurring in the tubes will be discussed in further detail below.

Tests were then undertaken on tubes with lengths between 10 and 40 mm (average D/t = 6) and typical force-displacement traces are shown in Figure 4. All four traces exhibit similar maximum force values, with the three shortest tubes offering values of approximately 26 kN and the longest withstanding a maximum force of 22.5 kN. Although these tubes are too short for buckling effects to intervene, it is likely that these length effects are associated with frictional effects between the plattens and the sample. In the shortest tubes, the presence of the constraint associated with the plattens is likely to extend over a significant length of the tube. As the tubes become longer, constraint effects due to the boundary conditions at the plattens play a smaller relative role, possibly leading to greater splitting and a drop in strength. Following the peak value, the force in all four traces drops steadily in a stable manner. It is clear that the maximum displacement increases with tube length, although these maximum values are much lower than the overall length of the tube. For example, the maximum displacement for the 40-mm-long tubes was approximately 10 mm, suggesting that the full energy-absorbing capability of the tubes has not been realized. The area under the load-displacement traces was then used to determine the variation in SEA with tube length and these values are shown in Figure 5. The average value of SEA for the shortest sample was 31.7 kJ/kg whereas as that for the longest specimens was only 12.8 kJ/kg. This clearly represents a significant variation in energy-absorbing characteristics and suggests that care has to be taken when selecting an appropriate length of reinforcement. This significant reduction in energy-absorbing capacity can be explained by examining the failed samples, typical examples of which are shown in Figure 6. In Figure 6(a), it is evident that the shortest sample exhibits a number of distinct failure mechanisms, including splits that propagate along the length of the tube, crushing of the uppermost part of the sample and cracking within the thickness of the bamboo tube. The 20-mm-long tube also exhibits fewer longitudinal splits in addition to top surface crushing. The SEA of this sample was only 60% of that shown in Figure 6(a), due to the fact that only one-third of this sample has been crushed, compared to approximately half in the 10-mm tube.

Mechanical properties of bamboo-reinforced foam

Any attempt to use bamboo tubes for energy absorption requires a low-density substrate to support them and maintain their position when in operation. The next part of this research study therefore focused on investigating the mechanical properties of foams reinforced with tubes. The density of the foam was varied in order to establish the influence of the degree of support applied to the outer surface of the tube on its energy-absorbing behavior. Tests were undertaken on 20-mm-long samples with values of D/t in the range 3.6 to 4.6. Although the previous section has shown that this is not the most effective geometry in terms of SEA, the use of larger D/t tubes does reduce the number of drilling and insertion operations required in multi-tube systems. Figure 7 shows typical load-displacement traces following compression tests on foams with different densities reinforced with a single bamboo tube. Also included is a load-displacement trace corresponding to a test on an individual unsupported bamboo tube. Increasing the foam density serves to increase both the peak load and the subsequent load-bearing capability of the structure. For example, the load response of the lowest density structure drops rapidly to values below 10 kN, whereas the lowest value for the 200 kg/m³ foam remains above 40 kN. Clearly, increasing the foam density serves to increase the overall amount of energy absorbed, as indicated by the greater area under the load-displacement trace. Figure 8 shows the variation of SEA with foam density, from where it is clear that this specific measure of energy absorption does not vary greatly with foam density, with all of the values being close to 20 kJ/kg. Attempts were then made to estimate the energy absorbed solely by the tubes, during which a series of tests were undertaken on foam blocks having a volume equal to that of the foam in a tube-reinforced sample (i.e. they were slightly smaller to account for the volume of the central hole). The energy absorbed in crushing the tube was then estimated by removing the contribution of the core from the energy under the load-displacement trace in the tube-foam combination. This value was then normalized by the mass for the reinforcement to yield an equivalent SEA for the bamboo tube. These values of SEA are included in Figure 8 from where it is evident that the energy absorbed by the bamboo tube increases rapidly with foam density, reaching a value of approximately 55 kJ/kg in the highest density system.

Figure 9 shows front and rear surface images of reinforced foams with nominal densities of 40 and 100 kg/m³. An examination of the front face of the low-density foam highlights the presence of several radial cracks, suggesting the presence of longitudinal splitting. Indeed, a significant number of these low-energy longitudinal splits can be seen in the image of the rear surface, Figure 9(a). The foam has clearly not prevented the splaying of the fractured lengths of tube during compression. In contrast, increasing the foam density to 100 kg/m³ serves to constrain the lateral movement of the tubular reinforcement, Figure 9(b). There is only a limited amount of splitting of the outer diameter of the tube in evidence. Closer inspection indicates that several of these cracks have extended into the adjacent foam. It is interesting to note that the ‘corrected’ SEA of the tube (i.e. following removal of the contribution of the foam) in this sample is 43 kJ/kg, suggesting that the foam has been successful in constraining the test sample and suppressing the longitudinal splitting failure mechanism.

Techniques for enhancing the energy-absorbing capacity of the bamboo tubes

The evidence presented above indicates that tube barreling and the associated longitudinal splitting are undesirable deformation and failure processes in these tubes, which should, if possible, be suppressed. It has been shown that embedding the fibers in a medium-density foam can reduce splitting, enabling higher energy modes to be triggered. Another option for suppressing this splitting mechanism is to wrap the tubes with a fiber reinforcement. In order to investigate this further, tubes of different length were wrapped with epoxy-coated kenaf fibers, Figure 10(a). The nominal weight fraction of fibers in the wrapped composite was 40%. Typical load-displacement traces fooling tests on 10-, 20-, 30- and 40-mm-long tubes (average D/t = 3.7) are shown in Figure 11. Comparing the traces in this figure with those in Figure 4 shows that the encapsulation process has a significant effect on the load-carrying capability of the tubes, serving to increase the maximum force by approximately 50% as well as greatly enhancing the post-peak load-bearing capacity of the structures. The resulting SEA values are compared with those following tests on the plain, unwrapped tubes in Figure 5 where it is evident that the reinforcement process serves to significantly increase the capability of the tubes. Figure 10(b) shows a failed test sample where it is evident that the failure process has changed from that apparent in the equivalent unwrapped tube, Figure 6(b). In this case, the reinforcing kenaf fibers appeared to have fractured in a tensile mode as the tube expanded under compression. The failed sample now exhibits a multiplicity of longitudinal splits, many of which may have occurred during the final stages of testing (i.e. following failure of the kenaf fibers). This evidence clearly supports the suggestion that suppressing gross barreling of the sample, by reinforcing the circumference of the tube, can modify the failure process and greatly enhance the crush performance of the tubes.

The evidence presented in this paper suggests that naturally occurring tube-like structures offer significant potential for use in environmentally friendly energy-absorbing structures. Future research will investigate multi-tube systems subjected to both impact testing and blast loading.

Conclusions

The energy-absorbing characteristics of a range of bamboo-reinforced foams have been investigated. Initial attention focused on assessing the crush characteristics of the individual foams, where it has been shown that the SEA decreases with increasing tube length. These trends were explained by investigating the failure mechanisms occurring during failure, where it was established the longer tubes failed predominantly in low-energy axial splitting mode whereas shorter tubes displayed significantly greater amounts of localized crushing, a mechanism that accounted for greater energy absorption.

The tubes were then embedded in a range of crosslinked PVC foams where it was demonstrated that higher density foams offered greater support to the tubes, resulting in a significant reduction in longitudinal splitting and increased levels of localized crushing. As a result, the energy absorbed by the individual bamboo tubes increased with foam density.

Finally, a number of tubes have been wrapped with a natural fiber reinforced epoxy resin in an attempt to enhance their hoop strength and reduce the level of longitudinal splitting. Here, it was noted that the wrapping procedure greatly reduces longitudinal splitting during compression resulting in increases in SEA. Future research will investigate the crush characteristics of environmentally friendly foams based on dense arrays of bamboo tubes.

Footnotes

Acknowledgements

The authors thank Airex AG (Switzerland) for kindly supplying the foams. Gratitude is also expressed to Mohd Zuhri Mohamed Yusoff for conducting a number of compression tests.

References

Thorton

Edwards

. Energy absorption in composite tubes. J Compos Mater 1982; 16: 521–545.

Melo

JDD

Silva

ALS

Villena

JEN

. The effect of processing conditions on the energy absorption capability of composite tubes. Compos Struct 2008; 82: 622–628.

Lau

STW

Said

Yaakob

. On the effect of geometrical designs and failure modes in composite axial crushing: a literature review. Compos Struct 2012; 94: 803–812.

Farley

Jones

. Prediction of the energy-absorption. J Compos Mater 1992; 26: 388–404.

Farley

. Energy absorption of composite materials. J Compos Mater 1983; 17: 267–279.

Jacob

Fellers

Simunovic

. Energy absorption in polymer composite materials for automotive crashworthiness. J Compos Mater 2002; 36: 813–850.

Song

H-W

X-W

Zhao

G-F

. Energy absorption behavior of double-chamfer triggered glass/epoxy circular tubes. J Compos Mater 2002; 36: 2183–2198.

Sigalas

Kumosa

Hull

. Trigger mechanisms in energy-absorbing glass cloth/epoxy tubes. Compos Sci Technol 1991; 40: 265–287.

Thuis

HGSJ

Metz

. Influence of trigger configurations and laminate lay-up on the failure mode of composite crush cylinders. Compos Struct 1993; 25: 37–43.

10.

Solaimurugan

Velmurugan

. Progressive crushing of stitched glass/polyester composite cylindrical shells. Compos Sci Technol 2007; 67: 422–437.

11.

Mamalis

Manolakos

Ioannidis

. On the response of thin-walled CFRP composite tubular components subjected to static and dynamic axial compressive loading: experimental. Compos Struct 2005; 69: 407–420.

12.

Hull

. A unified approach to progressive crushing of fibre-reinforced composite tubes. Compos Sci Technol 1991; 40: 377–421.

13.

Ramakrishna

Hamada

. Energy absorption characteristics of crashworthy structural composite materials. Key Eng Mater 1998; 141–143: 585–620.

14.

Turner

Warrior

Robitaille

. The influence of processing variables on the energy absorption of composite tubes. Compos Part A 2005; 36: 1291–1299.

15.

Mamalis

Manolakos

Demosthenous

. The static and dynamic axial crumbling of thin-walled fibreglass composite square tubes. Compos Part B 1997; 28: 439–451.

16.

Hamada

Ramakrishna

. Effect of fiber material on the energy absorption behavior of thermoplastic composite tubes. J Thermoplast Compos 1996; 9: 259–279.

17.

Farley

. Effect of specimen geometry on the energy absorption capability of composite materials. J Compos Mater 1986; 20: 390–400.

18.

Ruzanna A, Cantwell WJ, Langdon GS, et al. The energy-absorbing characteristics of composite tube-reinforced foam structures, submitted to Composite Structures (in press).

19.

. Impact tolerant and healable aluminum millitube reinforced shape memory polymer composite sandwich core. Mater Design 2013; 51: 79–87.

20.

Ouyang

. Debonding and impact tolerant sandwich panel with hybrid foam core. Compos Struct 2013; 103: 143–150.

21.

Zhang

Rong

. Fully biodegradable natural fiber composites from renewable resources: all-plant fiber composites. Compos Sci Technol 2005; 65: 2514–2525.

22.

Lim

J-K

. Fabrication and mechanical properties of completely biodegradable hemp fiber reinforced polylactic acid composites. J Compos Mater 2007; 41: 1655–1669.

23.

Herrara-Franco

Valadez-Gonzalez

. A study of the mechanical properties of short natural-fiber reinforced composites. Compos Part B 2006; 36: 597–608.

24.

Meredith

Ebsworth

Coles

. Natural fibre composite energy absorbing structures. Compos Sci Technol 2012; 72: 211–217.

25.

Mahdi

Hamouda

Sahari

. On the collapse of cotton/epoxy tubes under axial static loading. Appl Compos Mater 2003; 10: 67–84.

26.

Ataollahi

Taher

Eshkoor

. Energy absorption and failure response of silk/epoxy composite square tubes: experimental. Compos Part B 2012; 43: 542–548.

27.

Manufacturer’s data sheets for Airex C70 foams, Airex AG Switzerland.

28.

Osorio

Trujillo

Van Vuure

. Morphological aspects and mechanical properties of single bamboo fibers and flexural characterization of bamboo/epoxy composites. J Reinf Plast Compos 2011; 30: 396–408.

29.

Thorton

. Energy absorption in composite structures. J Compos Mater 1979; 13: 247–262.