Experimental and theoretical study of sandwich composites with Z-pins under quasi-static compression loading

Abstract

This study aims to enhance the compressive properties of sandwich composites containing extruded polystyrene (XPS) foam core and glass or carbon face materials by using carbon/vinyl ester and glass/vinyl ester composite Z-pins. The composite pins were inserted into foam cores at two different densities (15 and 30 mm). Compression test results showed that compressive strength, modulus and loads of the sandwich composites significantly increased after using composite Z-pins. Sandwich composites with 15 mm pin densities exhibited higher compressive properties than that of 30 mm pin densities. The pin type played a critical role whilst carbon pin reinforced sandwich composites had higher compressive properties compared to glass pin reinforced sandwich composites. Finite element analysis (FE) using Abaqus software has been established in this study to verify the experimental results. Experimental and numerical results based on the capabilities of the sandwich composites to capture the mechanical behaviour and the damage failure modes were conducted and showed a good agreement between them.

Keywords

compression strength glass and carbon fibres sandwich composites XPS foam Z-pins

Introduction

Sandwich composites made from high performance fibre laminate faces over foam cores are highly used in aerospace, naval applications and wind turbines where the weight is critical (Carlsson and Kardomateas, 2011; Lal and Markad, 2020a, 2020b; Selver and Kaya, 2019b). The core part plays critical role to provide compressive strength to sandwich composites whilst it is weaker than face or skin materials (Thomsen et al., 2005). In recent years, there has been an increasing amount of literature on studying compressive properties of sandwich composites using different types of core materials (Liu et al., 2018; Mouritz, 2006; Xiong et al., 2010). Avery and Sankar (2000) analysed compressive properties of sandwich composites made of graphite/epoxy face-sheets and aramid fibre honeycomb core.Malcom et al. (2013) investigated the compressive properties of Divinycell core/glass fibre sandwich composite structures reinforced with stitching of corrugated core using Kevlar yarns. Lascoup et al. (2006) studied the effect of angular stitching on bending and compression properties sandwich composed of a foam core and two thin woven glass fibre skins. Lainé et al. (2013) manufactured and investigated compressive behaviour of sandwich composites made from polymeric foam core reinforced by Napcos technology based on transverse needling. Dai et al. (2016) developed and studied the compressive strength of looped fabric reinforced foam core sandwich composites.

There is a large volume of published studies describing the role of reinforcing core parts on compressive properties of sandwich composites. Mei et al. (2017) observed that sandwich composite panels made from tetrahedral truss core exhibited higher specific compressive strength than that of metallic truss core sandwich panels. Wang et al. (2010) embedded carbon rods into pyramidal lattice truss core part of the sandwich composites to investigate the buckling and shear performance under compressive loading. Yalkin et al. (2015) attempted to enhance the compressive behaviour of polyvinyl chloride (PVC) foam core and glass face composites using glass fibre stitching through the thickness direction. They observed that compressive strength increased with increasing stitching yarn thickness. Potluri et al. (2003) reinforced Divinycell core/glass face sandwich composites using various stitching space (5–25 mm).

Recent developments in the field of through-thickness reinforcement of composites have led to a renewed interest in using Z-pinning technology (Cartie and Fleck, 2003; Kaya and Selver, 2019; Li et al., 2012; Long et al., 2008; Marasco et al., 2006; Selver and Kaya, 2019a; Wallace et al., 2001). Jayaram et al. (2019) observed that using polyester resin pins enhanced compressive load bearing capacity of foam filled honeycomb sandwich panel. Abdi et al. (2014) reinforced foam core sandwich composite panels with 2 and 3 mm polyester pins and observed that increasing the pins diameter improved the flexural and flat-wise compression strength. Long et al. (2008) studied the effects of Z-pin configurations (inclination angle and pinning density) on damage modes of foam core sandwich composite structures. Lei et al. (2016) analysed buckling and crushing behaviour of foam-core/glass face sandwich composites after reinforcing with resin pins. They observed that using pins enhanced compressive properties of sandwich panels. Nanayakkara et al. (2011) used unidirectional thin and thick carbon filament rods to improve compressive properties of sandwich composite made from carbon-epoxy laminate face skin and low density core of polymer foam. Virakthi et al. (2018) modelled out-of-plane compressive, shear stiffness and strength of X-Cor sandwich composites using aligned pins in pyramidal geometry. They claimed that foam core increases the strength of sandwich composites due to postponing buckling of pins.

The studies presented thus far provide evidence that using Z-pins can enhance the compressive properties of sandwich composites. However, few studies have investigated the compressive properties of sandwich structures consisted of different types of composite pins or rods. Also, there is little published data on compressive properties of sandwich composites consisted of different face materials and reinforcement pins with different arrangements. Therefore, the aim of this paper is to explore the effect of Z-pin types on compressive properties of XPS foam core sandwich composites using glass/vinyl ester and carbon/vinyl ester pins with two different rod/pin densities (15 and 30 mm). The compression behaviour of sandwich composite was predicted theoretically using finite element analysis with aiding Abaqus software. The numerical results based on the mechanical behaviour and damage failure mode were compared with that obtained from experimental results.

Materials and methods

Materials

Sandwich composite panels were manufactured from woven glass or carbon face materials with [0/90]₃ configurations and a XPS foam core. Areal densities of glass and carbon fabrics are 280 and 245 g/m², respectively and purchased from DostKimya. Core part of the sandwich panels were reinforced with 3 mm diameter of carbon/vinyl ester and glass/vinyl ester composite pins (Table 1) using 15 and 30 mm pin distances by manually (Figure 1(a)) purchased from (DostKimya, 2018). Glass and carbon pins are made of unidirectional glass or carbon fibres infused with vinyl ester matrix material. The holes for the pins were pre-drilled using metal rods with a similar diameter of carbon and glass pins. Then, the carbon/glass pins were placed on those pre-drilled holes manually. Face fabric materials were placed over and under the XPS foams with [0/90]₃ configurations to prepare the sandwich preforms. Sandwich composite preforms were infused with L160 epoxy resin and H160 hardener (Hexion MGS) using vacuum bagging method (Figure 1(b)) at 50°C for 1 h. Sandwich composites and their face materials (GF and CF) are presented in Table 2 with their sample codes. Figure 1(c) and (d) presents low (30 mm) and high (15 mm) density pin placement. Figure 2 shows Z-pin reinforced sandwich composites after removing the core part using ethyl acetate (30 min) at room temperature. It can be seen that there are still remaining XPS parts on the surface of the carbon pins in contrast to glass pins. This is due to carbon pins showed better heat transfer compared to glass pins during the curing process and the XPS foam core bonded to carbon pins better than glass pins.

Table 1.

Properties of composite pins (DostKimya, 2018).

	Glass/vinyl ester	Carbon/vinyl ester
Fibre volumefraction (%)	65	65
Tensilestrength(MPa)	1000	2080
Compressionstrength(MPa)	550	1450
Tensilemodulus(GPa)	38	145
Density (g/cm³)	1.8	1.5
Cross-sectiondiameter (mm)	3	3

Table 2.

Sandwich composite samples.

Sample code	Face	Core	Pin type	Pin density (mm)	Composite density (g/cm³)
GF	Glass	–	–	–	1.85 (±0.01)
CF	Carbon	–	–	–	1.48 (±0.01)
GS	Glass	XPS	–	–	0.37 (±0.004)
CS	Carbon	XPS	–	–	0.35 (±0.014)
GS-C15	Glass	XPS	Carbon	15	0.61 (±0.004)
GS-C30	Glass	XPS	Carbon	30	0.41 (±0.008)
GS-G15	Glass	XPS	Glass	15	0.69 (±0.007)
GS-G30	Glass	XPS	Glass	30	0.46 (±0.003)
CS-G15	Carbon	XPS	Glass	15	0.43 (±0.005)
CS-G30	Carbon	XPS	Glass	30	0.39 (±0.001)
CS-C15	Carbon	XPS	Carbon	15	0.42 (±0.011)
CS-C30	Carbon	XPS	Carbon	30	0.40 (±0.011)

Figure 1.

Images of carbon pin placement to XPS (a), vacuum infusion method (b), placement of loose (c) and dense (d) pins.

Figure 2.

Images of sandwich composites before and after removing the core parts: (a, b) CS-G15, (c) CS-C30, (d, e) GS-G30, and (f) GS-C30.

Methods

ASTM D792-08 standard was used to calculate the densities of face and sandwich composites using a digital densimeter. Fibre volume fraction of glass/epoxy face composites (GF) was measured with BS EN ISO 1172:1999 standards whilst volume fraction of carbon/epoxy face composites (CF) was calculated theoretically due to burning problem of carbon fibres during the measurement using equation (1):

\begin{matrix} Volume fraction (Vf) = \frac{Volume of fiber}{Volume of composite} = \frac{\frac{Wf}{df}}{L . w . h} \\ = \frac{\frac{nlayer . gr . L . w}{df}}{L . w . h} = \frac{nlayer . gr}{h . df} \end{matrix}

(1)

Where V_f, reinforcement fibre volume fraction (%); W_f, weight of reinforcement fibre; df, density of fibre; L, length of the specimen; w, width of the specimen; h, thickness of the specimen; gr, areal density of the fabric; nlayer, number of the layer in the laminate (Cherif et al., 2013).

The thickness of sandwich composites varied between 20.0 and 21.9 mm. The thickness of face sheets of GF and CF were 1.48 and 1.59 mm, respectively. Fibre volume fraction of face sheets of GF and CF were 47.47% and 51.37%, respectively. The density results of sandwich composites are presented in Table 2. The densities of face sheets were 1.85 g/cm³ for GF and 1.48 g/cm³ for CF. It can be seen that the density of sandwich composites (GS and CS) increased after inserting of glass and carbon pins. It is clear that composites with 15 mm pin placement (GS-C15, GS-G15, CS-C15 and CS-G15) have higher densities than that of 30 mm pin placements (GS-C30, GS-G30, CS-C30 and CS-G30). It is also clear that using glass pins and glass face materials resulted in higher density values compared to using carbon pins and carbon face materials.

Sandwich composite specimens were cut into 50 × 50 mm, and they were tested using a ZwickRoell/Z100 testing machine at a constant displacement of 0.5 mm/min according to ASTM C365/C365M-16 standard test method for compression test. Four specimens were tested for each sample. The final specimens have 4 pins and 9 pins for low (GS-C30, GS-G30, CS-G30, CS-C30) and high (GS-C15, GS-G15, CS-G15, CS-C15) density pin reinforced sandwich composite samples, respectively. Failures of z-pins after compression test were examined by using an optical microscope (BAB Bs200Doc, Turkey).

Numerical model

In order to assess whether the design, assumption, concerning material model’s selection and material constant experimental determination are correct, numerical simulations of quasi-static compression loading of composite sandwich with foam and pins are described in detail. The discrete models are built in Abaqus 6.14 Finite Element Method (FEM) environment. Based on the dimension are given in the experimental, applied material solution, and observed failure mechanism, the model variants are considered. The following sections will discuss the FEM models.

Material models

This model represents pure laminate-foam-laminate structure as shown in Figure 3(a). Thus, the orthotropic properties of laminates are simulated using the elastic type (i.e. engineering constant), meanwhile, the isotropic properties of XPS foam are simulated using elastic type (i.e. Isotropic). Further, the plasticity model and crushable foam hardening was adopted to simulate the plastic hardening of the cores. Two types of composites which are glass and carbon laminates were used for face sheet materials. The properties of glass and carbon, and XPS foam are illustrated in Table 3. Additionally, glass and carbon composites pins are included between face sheets, so small holes in the foam (∼3 mm) are made which allow resin to flow through it and form both laminate skins. It is believed that their consideration will improve results accuracy of obtained response sandwich composites. The height of pins is same as the height of foam (∼20 mm), and the space distance between the adjacent columns (pins) are 15 mm and 30 mm respectively as illustrated in Figure 3(b) and (c). Two types of materials, which used for manufacturing these pins in this study, the former is manufactured from carbon/vinyl ester, and the latter is made from glass/vinyl ester. The specification of these pins is also shown in Table 3.

Table 3.

Elastic and mechanical properties of composite laminates.

Property	Glass face sheet	Carbon face sheet	Glass pins	Carbon pins	XPS foam
t (mm)	1.48	1.59	20	20	20
$ρ (\frac{K g}{m^{3}})$	1850	1480	1800	1500	33
$E_{11} (GPa)$	8.750	14.6	38	145	0.0167
$E_{22} (GPa$ )	7.058	14	38	140	0.0157
$E_{33} (GPa)$	7.058	7.058	30	145	0.0160
$G_{12} (GPa)$	2.17	2.16	1.05	2.142
$G_{23} (GPa)$	2.71	2.36	1.01	2.142
$G_{13} (GPa)$	2.17	2.16	1.05	3.120
$ν_{12}$	0.29	0.26	0.35	0.208	0.35
$ν_{23}$	0.30	0.28	0.36	0.208	0.35
$ν_{13}$	0.29	0.26	0.35	0.250	0.35
$X_{t} (MPa$ )	450	830	70.4	2080	0.120
$X_{c} (MPa)$	133	600	35,53	2000	0.100
$Y_{t} (MPa)$	400	800	27.84	1450	0.347
$Y_{c} (MPa)$	80.30	500	93.16	800	0.179
$Z_{t} (MPa) *$	73	73	–
$S_{12} (MPa)$	223	300	35	60
$S_{23} (MPa)$	225	280	38	55
$S_{13} (MPa)$	223	300	35	55

Figure 3.

The FEM model for sandwich composites: (a) XSP foam and face sheet, (b) XSP foam, face sheet, and 4 pins, and(c) XSP foam, face sheet and 9 pins.

The failure of the face sheet of sandwich composites simulated by adoption the continuum damage initiation and damage evolution based on the Hashin Damage Criteria. This criterion consists of four different criteria; tensile fibre failure $F_{t 1}$ , compressive fibre failure $F_{1 c},$ tensile matrix failure $F_{2 t}$ and compressive matrix failure $F_{2 c}$ . Thus, the Hashin criteria for the two-dimensional case are presents as follow,

Tensile fibre failure F_{1 t} = {(\frac{σ_{11}}{X_{t}})}^{2} if σ_{11} ⩾ 0

(2)

Compressive fibre failure F_{1 c} = {(\frac{σ_{11}}{X_{c}})}^{2} if σ_{11} < 0

(3)

Tensile matrix failure F_{2 t} = {(\frac{σ_{22}}{Y_{t}})}^{2} + {(\frac{τ_{12}}{Z_{l}})}^{2} if σ_{22} > 0

(4)

\begin{matrix} Compressive matrix failure F_{2 c} = {(\frac{σ_{22}}{Y_{c}})}^{2} + {(\frac{τ_{12}}{Z_{l}})}^{2} \\ if σ_{22} < 0 \end{matrix}

(5)

Where, in the equations (2)–(5), the quantities $X_{t}$ and $X_{c}$ denote the longitudinal tensile and compressive strength, respectively, $Y_{t}$ and $Y_{c}$ denote the transverse tensile and compressive strength, respectively, $Z_{l}$ denotes the longitudinal shear strength, $σ$ is the stress which applied on the element along various directions of composite, and $τ$ stands for shear strength of the element.

The crushable foam hardening which is available in Abaqus used to model XPS foam materials in order to simulate the plastic hardening of XPS foam. Thus, the isotropic properties are used, and the materials plasticity values of stress are required.

The XPS foam behaved elastically up to a certain point (yield point or point a in Figure 4), when compressed, after this yield point, it behaved plastically (resulting in a plateau) up to a certain point (point b in Figure 4). Compressing it even further results in densification of the XPS foam, which means that the small cells in the foam are packed together until the foam fails (point c in Figure 4). This behaviour can be clearly seen when the stress-strain curve is plotted during a compression test of XPS foam as seen in Figure 4.

Figure 4.

Stress-strain curve of XPS foam in a compression test.

In this study, the crushable XPS foam parameters are defined based on its properties, which presented in Table 3. In addition, the crushable foam hardening parameters (i.e. Yield stress-plastic strain data) are extracted from compression load-strain graph of XPS foam as illustrated in Figure 4. The static situation is preformed to simulate the compression loading of sandwich composites and period of time that selected for this simulation is 0.001 s.

Mesh type

The mesh of the face sheets and pins are standard type with continuum shell elements (SC8R) with linear geometric order. Further, the element type used is structural hex, which stacked from the top plane. While the mesh of the XPS foam is standard type (3D stress) elements (C3D8R) with linear geometric order and the element type that is used is structural hex stacked from the top plane.

Interactions

This simulation has surface-to-surface contact between the face sheets and fixtures with finite sliding, since the fixtures are considered, as the master surfaces and the faces sheet are the slave surface and this contact is modelled as a hard contact. Additionally, the face sheet and XPS core are bonded together using tie interaction in order to ensure their rigid bonding. This means that the slave surface makes the exact same movement as the master surface at each mode. Because of the load presses from the top, it is decided that the upper fixture is the master surface and the faces sheet underneath is the slave surface and for bottom surface pair the bottom fixture is the master surface and the skin is the slave surface.

Loading and boundary conditions

The models are loaded by upper face sheet (purple colour surface as shown in Figure 5(a)) with a displacement 2.5 mm in vertical direction and represented by Z-direction. Further, the boundary conditions of this simulation are shown in Figure 5(b), which can be seen that the face sheet in bottom (purple colour surface) is encastered and that provide a fixed for all degree of freedom. Moreover, the sandwich composite is imposed with X and Y-symmetry as shown in Figure 5(c) (red edges and purple base). The process of this simulation is illustrated in Figure 6.

Figure 5.

Loading and boundary condition of the FEM compression models of sandwich composites: (a) applying load on the top (purple colour), (b) fixing the bottom part (purple colour), and (c) edges are imposed with X and Y-symmetry (purple and red colours).

Figure 6.

Flow chart of the process of the simulation.

Results and discussions

Experimental results

Table 4 presents compression test results of XPS core and sandwich composites with and without pin reinforcement. It can be seen that XPS core foam had lower compressive load and stress than that of sandwich composites. Comparing sandwich composites, carbon/XPS sandwich composite (CS) had slightly higher compressive load and strength than that of glass/XPS sandwich composite (GS). Figure 7 compares stress-displacement relation of the unpinned sandwich composites. The compressive strength values are the peak points where the stress values start to drop in the plots. Although GS sandwich composites had lower strength, they had higher extension values than CS sandwich composites. Table 4 presents that addition of carbon and glass Z-pins significantly enhanced compressive loads of GS and CS sandwich composites. This is due to Z-pins are much stiffer and stronger than the XPS foam core. It can be seen that compressive loads of GS sandwich composites increased from 0.568 to 14.04 kN and 14.88 kN after using glass face with glass and carbon pins at 30 mm spacing (GS-G30 and GS-C30). Sandwich composites with denser pin placements (GS-G15 and GS-C15) showed about 120-145% higher compressive loads than their lower density pin placements (GS-G30 and GS-C30). It is clear that there are no significant differences between using glass or carbon rods for compressive loads of glass face sandwich composites. For carbon face composites, using carbon pins slightly increased compressive loads of sandwich composites compared to using glass pins. Table 4 states that compressive loads of CS-C15 composites are about 18% higher than CS-G15 composites whilst compressive loads of CS-C30 and CS-G30 samples are very similar. This indicates that using carbon pins at higher densities (15 mm) are more effective than using them at low densities (30 mm) compared to glass pins.

Table 4.

Compressive properties of sandwich composites.

Sample	Compressiveload-experimental(kN)	Compressiveload-numerical(kN)	Experimentaland numericaldifferences(load %)	Compressive strength(MPa)	NormalizedCompressivestrength(MPa)
XPS	0.309 (0.030)	–		0.123 (0.04)	–
GS	0.568 (±0.03)	0.615	+ 8.25	0.185 (±0.02)	–
CS	0.615 (±0.01)	0.700	+ 13.73	0.207 (±0.01)	–
GS-G30	14.04 (±1.1)	14.00	−0.36	5.61 (±0.5)	1.40
GS-G15	31.52 (±1.0)	32.81	+4.08	12.32 (±1.0)	1.36
GS-C30	14.88 (±1.9)	14.22	−4.50	5.80 (±0.6)	1.45
GS-C15	36.36 (±4.5)	35.35	−2.76	12.72 (±2.4)	1.41
CS-G30	18.28 (±1.7)	18.00	−1.55	7.31 (±1.4)	1.82
CS-G15	28.69 (±1.1)	30.50	+6.31	11.79 (±1.3)	1.31
CS-C30	18.11 (±0.1)	18.00	−0.63	7.20 (±0.2)	1.8
CS-C15	33.98 (±1.5)	35.35	+4.05	12.88 (±0.7)	1.43

Figure 7.

Compressive stress and strain history of carbon (CS) and glass (GS) sandwich composites.

Table 4 also presents the compression strength values, which are normalised to number of Z-pins for each sample. It can be seen that samples with glass faces had very similar normalised compression strength values. However, carbon face sandwich composites with low-density (CS-C30 and CS-G30) pin placements had higher normalised values than their higher pin placements (CS-C15 and CS-G15). This is due to placing pins with lower density lead to less contact points between face materials and pins. Hence, number of debonding reduced and samples carried more loads compared to samples with more contact points.

Figure 8 presents compressive stress-strain history of carbon and glass pins reinforced sandwich composites. Table 4 indicates that addition of carbon and glass pins significantly enhanced compressive strength of glass face sandwich composites. It can be seen that compressive strength of glass and carbon pin reinforced sandwich composites (GS-C30 and GS-G30) are very similar when the pin density is low. What is interesting in this data is that compressive strength of carbon pins are about three times higher than that of glass pins according to Table 1 although they showed similar results when they were used as a part of sandwich composites. This might be due to the XPS foams inhibits or postpones buckling of pins and they can handle higher compressive loads for glass pin reinforced composites. It seems possible that the interaction between XPS foam core and pins lead to fibre kinking for glass pins under compressive loads which can be seen in Figure 9. These kinking mechanisms can inhibit misalignment of fibres inside the pins and provide advantages by delaying compressive failure for the sandwich composites. However, carbon pin reinforced sandwich composites exhibit slightly higher compressive strength than glass pin reinforced sandwich composites when the pin density is high as for GS-C15 and GS-G15. A possible explanation for this is that compressive strength of carbon pins is higher than glass pins as presented in Table 1. It is also clear that compressive stress increases with increasing pin densities for glass face sandwich composites.

Figure 8.

Compressive stress and displacement history of: (a) glass face and (b) carbon face sandwich composites.

Figure 9.

Optical microscope images of z-pin failures after compression tests for carbon face: (a) CS-C15, (b) CS-C30, (c) CS-G15, (d) CS-G30 and glass face: (e) GS-C15, (f) GS-C30, (g) GS-G15, and (h) GS-G30 composites.

Figure 8(a) indicates that compressive stress some of the samples undulates with multiple (four) breaking points under compressive load. For instance, one of the glass pins fails first at around 1 mm displacement for GS-G30 sample. Then, it still continues to carry the compressive load until 1.3 mm displacement value before failing again. Breaking of the all glass pins can be seen for GS-G30 sample. On the other hand, some of the samples show less breaking point than they contain (four pins for low and nine pins for high density arrangement). This is due to some of the pins fail together during the compressive loading in contrast to individual pin failures. Similarly, carbon face sandwich composites also showed multiple breaking points as shown in Figure 8(b).

Table 4 and Figure 8(b) presents stress-displacement relation of carbon face sandwich composites during compressive loading. It can be seen from Table 4 that addition of glass or carbon Z-pins significantly increased compressive strength of the neat carbon face sandwich composite (CS). It is apparent from Figure 8(b) that compressive strength of carbon sandwich composite with high-density glass pins (CS-G15) is about 73% higher than that of sandwich composites with low-density glass pins (CS-G30). What is interesting about the data in this table is that sandwich composites with high-density carbon pins (CS-C15) exhibits about 79% higher strength values than low-density carbon pins (CS-C30). Comparing pin types for carbon face composites, sandwich composites with carbon pins had about 5.5% higher compressive strength than sandwich composites with glass pins at low-density pin arrangement. Similar results were achieved when the pins were placed with high density whilst carbon pins reinforced composites exhibited about 9.2 % higher strength values than glass pin reinforced sandwich composites.

Comparing glass and carbon face composites, carbon face sandwich composite with low-density glass pins (CS-G30) had about 21% higher compressive strength than glass face sandwich composites with low-density glass pins (GS-G30). However, carbon face sandwich with high-density glass pins (CS-G15) exhibited about 5% lower values than glass face sandwich composites with high-density glass pins (GS-G15). This reduction is due to misaligned Z-pins during manufacturing of sandwich composites; hence, they fail earlier than their aligned forms. For carbon pin comparison, carbon face composite (CS-C30) had about 24% higher than glass face composite (GS-C30) at low-density arrangement. The observed increase in compressive strength could be attributed to bonding of carbon faces or carbon pins with XPS are better than glass faces, or glass pins as observed in Figure 2. It is apparent that there are still remaining XPS parts over the carbon pins after the ethyl acetate treatments, indicating that carbon fibres make better contact or bonding with XPS compared to glass fibre. This better bonding increases the compressive loads due to postponing buckling or crushing of Z-pins during compression test. On the other hand, carbon and glass faces composites with high density pins (CS-C15 and CS-G15) had slightly different values whilst CS-C15 composite had about 10% higher compressive strength than that of CS-G15 composite due to better carbon face/carbon pin bonding compared to glass face/carbon pin bonding.

Figure 9 presents the optical microscope images of fracture types of Z-pins for carbon face composites after compressive loading. It can be seen that carbon and glass pins showed slightly different failure mechanism. Carbon pins mainly failed with crushing and fibre fracture through the loading direction. However, glass pins fractured by splitting with fibre kinking. These pin splitting might be the reason why glass pin reinforced sandwich composites had lower compressive properties than carbon pin reinforced sandwich composites. It can be seen that carbon pins were fractured or crushed for glass face sandwich composites during compressive loading. However, glass pins were splitted and crushed for GS-C15 and GS-C30 samples, respectively. These show that the pin types affect the failure types and compressive behaviour of sandwich composites.

Figure 10(a) presents the amount of energy absorbed that was determined by integrating the area under the compressive stress-strain curve of sandwich composites. It is apparent that samples with higher pin densities had higher energy absorption than lower pin densities. What is striking about the figure is using carbon pins lead to absorb more energy than using glass pins. This is probably due to carbon pins had higher yielding strain during the compressive loading and they failed with buckling rather than crushing or splitting as observed for glass pins (as seen in Figure 9). The energy absorption capacity per unit weight was also evaluated in Figure 10(b) as previous researchers (Yang et al., 2020). The absorbed energy values under the stress-strain curves were divided by the density of composites (Table 2) to calculate the absorption capacity of samples per unit weight. It can be clearly seen that samples with higher pin gaps (15 mm) had higher energy absorption capacities than that of lower pin gaps (30 mm) as also observed in Figure 10(a). The only exception is CS-C30 and CS-C15 sandwich composites where they have very close energy absorption capacities.

Figure 10.

Energy absorption of sandwich composites: (a) absorbed energy and (b) absorbed energy per weight.

Numerical model validation

The numerical model, which adopted in this study, was validated by comparing its prediction to experimental data that obtained from compression tests, which applied on the sandwich composites. Figure 11 shows the experimental and numerical compression load verse displacement for the sandwich composite laminates without pins (i.e. CS and GS samples).

Figure 11.

Compression load-displacement graphs comparison between modelling and experimental tests for: (a) CS and (b) GS.

It can be seen that a good agreement between experimental and numerical results for both types of sandwich composites in the elastic region (∼ 0 to 550 N). However, the maximum forces in experimental data do not match with that in numerical results. This can be explained by poor adhesion between skin and XPS core, so that, wrinkling rupture initially started, then shearing and delamination have been occurred resulted in maximum force reduction (Kwon et al., 1995) meanwhile the modelling is based on the assumption that there is perfect bonding between skin and XPS core in sandwich composites.

The effect of the face sheet (i.e. glass and carbon composite laminates), the type, and numbers of the Z-pins on the compression behaviour of the sandwich composites are illustrated in Figure 12. Figure 12(a) to (d) shows the influence of types and numbers of Z-pins on the compression behaviour of sandwich composites with face sheet made from glass fibre laminates and XPS core. In Figure 12, the experimental and numerical results revealed a good agreement between them and the sandwich composites with higher numbers of Z-pins (i.e.nine pins) as seen in Figure 12(b) and (d) had a more compression resistance than the sandwich composites had lower numbers of Z-pins (i.e. four pins) as seen in Figure 12(a) and (c). This indicates that increasing of pin density through-thickness of sandwich composites can play role for enhancing their performance under compression loading. However, the sandwich composites having four or nine pins made from carbon composites exhibited slightly higher compression resistance than the sandwich composites having pins made from glass composites. This is because of the higher mechanical properties of pins made of carbon composites (as seen in Table 1).

Figure 12.

Compression load-displacement graphs comparison between modelling and experimental tests for: (a) GS-G30,(b) GS-G15, (c) GS-C30, (d) GS-C15, (e) CS-G30, (f) CS-G15, (g) CS-C30, and (h) CS-C15.

The compression resistance of sandwich composite has been influenced with mechanical properties of skin as be noted in Figure 12(e) to (h). According to the experimental and numerical results of compression load-displacement, it should be noted that the sandwich composite with carbon skin and 4 pins made of either glass or carbon composites appeared higher compression resistance than sandwich composite with glass skin and 4 pins as seen in Figure 12(e) to (h). With an increasing the density of Z-pins (i.e. nine pins), the compressive load capacity of both sandwich composites kept stable. In addition, maximum compressive force (N) of composites are compared experimentally and numerically and presented in Table 4 in which the difference values are compared and expressed in percentage difference. It seems that the modelling and experimental results show good agreements.

The damage failures of sandwich composites without Z-pins (skin and XPS foam) are presented in Figure 13. It can be clearly seen, the failure of this composite occurred when the skin debonded from the XPS foam. The Von-Misses stress criteria demonstrated that the XPS foam was damaged before the face sheet (i.e. glass or carbon laminates) in sandwich composite.

Figure 13.

The Von-Misses failure criteria for sandwich composite without pins (skin and XPS foam only) based on the numerical simulation.

Furthermore, the failure mechanism of sandwich composite having Z-pins (skin, XPS foam, and pins) are shown in Figure 14. It can be noted that the XPS foam started to damage surrounding pins as seen in Figure 14(a) causing pins-XPS foam debonding. Consequently, these damages produced cracks, which are initiated and propagated to reach the interfaces between skin and XPS core leading to skin-XPS foam debonding as seen in Figure 14(b).

Figure 14.

The Von- Misses failure criteria for sandwich composite with pins based on the numerical simulation (a), the debonding of skin- XPS foam and pins-XPS foam in sandwich composite based on the numerical simulation (b).

Conclusions

The main goal of the current study was to determine the effect of carbon and glass Z-pins on compressive properties of sandwich composites made from XPS foam core and glass or carbon fabric face materials. The following conclusions can be drawn from this study:

Compressive strength, modulus and loads of the sandwich composites were increased by addition of carbon or glass Z-pins which were much stiffer and stronger than the XPS foam core.

Sandwich composites with carbon face materials had mostly higher compressive properties than sandwich composites with glass face materials after and before the Z-pin reinforcements due to higher stiffness of carbon face compared to glass face composites.

Sandwich composites with high density pin placement showed higher compressive strength than low density pin placement.

Using carbon pins enhanced compressive properties of sandwich composites more than using glass pins at both low- and high-density pin placement because of the compressive strength of carbon pins are higher than glass pins.

Energy absorption of composites increased with increasing pin densities whilst samples with carbon pins exhibited higher energy absorption values than that of glass pins.

The experimental and simulated response of sandwich composites with and without pins are shown a good matching and it demonstrated the XSP foam model, faces sheet-XSP foam, and face sheet-pins interactions are represented well. The results also illustrated that the FE analysis can predict the compression behaviour of sandwich composites. However, the small discrepancy observed and that could be attributed to the simulation assumed a perfect bonding between face sheet and XSP foam and Z-pins respectively.

Footnotes

Declaration of conflicting interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by Kahramanmaras Sutcu Imam University Scientific Research Unit under Grant number 2016/6-57M.

ORCID iD

Erdem Selver

References

Abdi

Azwan

Abdullah

, et al. (2014) Flatwise compression and flexural behavior of foam core and polymer pin-reinforced foam core composite sandwich panels. International Journal of Mechanical Sciences 88: 138–144.

Avery

Sankar

(2000) Compressive failure of sandwich beams with debonded face-sheets, Journal of Composite Materials 34(14): 1176–1199.

Carlsson

Kardomateas

(2011) Structural and Failure Mechanics of Sandwich Composites. Dordrecht, The Netherlands: Springer.

Cartie

Fleck

(2003) The effect of pin reinforcement upon the through-thickness compressive strength of foam-cored sandwich panels. Composites Science and Technology 63(16): 2401–2409.

Cherif

Poilâne

Falher

, et al. (2013) Influence of textile treatment on mechanical and sorption properties of flax/epoxy composites. Polymer Composites 34(10): 1761–1773.

Dai

Zhou

Sun

, et al. (2016) Experimental study on the mechanical properties of looped fabric reinforced foam core sandwich composite. Journal of Composite Materials 50(20): 2807–2821.

DostKimya (2018) Kompozit.net. Available at: http://www.kompozit.net/ (accessed 20 January 2018).

Jayaram

Nagarajan

Vinod Kumar

(2019) Compression and low velocity impact response of sandwich panels with polyester pin-reinforced foam filled honeycomb core. Journal of Sandwich Structures and Materials 21(6): 2014–2030.

Kaya

Selver

(2019) Impact resistance of Z-pin-reinforced sandwich composites. Journal of Composite Materials 53(26–27): 3681–3699.

10.

Kwon

Murphy

Castelli

(1995) Buckling of unbalanced, sandwich panels with titanium and GRP skins. Journal of Pressure Vessel Technology 117(1): 40–44.

11.

Lainé

Le Grognec

Comas Cardona

, et al. (2013) Analytical, numerical and experimental study of the bifurcation and collapse behavior of a 3D reinforced sandwich structure under through-thickness compression. International Journal of Mechanical Sciences 67: 42–52.

12.

Lal

Markad

(2020a) Influence of dynamic temperature variation and inplane varying loads over post-buckling and free vibration analysis of sandwich composite beam. International Journal of Computational Materials Science and Engineering 9(3): 2050012.

13.

Lal

Markad

(2020b) Nonlinear flexural analysis of sandwich beam with multi walled carbon nanotube reinforced composite sheet under thermo-mechanical loading. Curved and Layered Structures 7(1): 1–16.

14.

Lascoup

Aboura

Khellil

, et al. (2006) On the mechanical effect of stitch addition in sandwich panel. Composites Science and Technology 66(10): 1385–1398.

15.

Lei

Yao

Wen

, et al. (2016) Experimental and numerical investigation on the crushing behavior of sandwich composite under edgewise compression loading. Composites Part B: Engineering 94: 34–44.

16.

Yan

Wang

, et al. (2012) Study on compressive properties of Z-pinned laminates in RTD and hygrothermal environment. Chinese Journal of Aeronautics 25(1): 64–70.

17.

Liu

Mei

, et al. (2018) Investigation on manufacturing and mechanical behavior of all-composite sandwich structure with Y-shaped cores. Composites Science and Technology 159: 87–102.

18.

Long

Guiqiong

Tao

(2008) Investigation of the Effect of Z-Pin Reinforcement on the Collapse of Foam-Cored Sandwich Panels. Journal of Reinforced Plastics and Composites 27(11): 1211–1224.

19.

Malcom

Aronson

Deshpande

, et al. (2013) Compressive response of glass fiber composite sandwich structures. Composites Part A: Applied Science and Manufacturing 54: 88–97.

20.

Marasco

Cartié

DDR

Partridge

, et al. (2006) Mechanical properties balance in novel Z-pinned sandwich panels: Out-of-plane properties. Composites Part A: Applied Science and Manufacturing 37(2): 295–302.

21.

Mei

Liu

(2017) A novel fabrication method and mechanical behavior of all-composite tetrahedral truss core sandwich panel. Composites Part A: Applied Science and Manufacturing 102: 28–39.

22.

Mouritz

(2006) Compression properties of z-pinned sandwich composites. Journal of Materials Science 41(17): 5771–5774.

23.

Nanayakkara

Feih

Mouritz

(2011) Experimental analysis of the through-thickness compression properties of z-pinned sandwich composites. Composites Part A: Applied Science and Manufacturing 42(11): 1673–1680.

24.

Potluri

Kusak

Reddy

(2003) Novel stitch-bonded sandwich composite structures. Composite Structures 59(2): 251–259.

25.

Selver

Kaya

(2019a) Flexural properties of sandwich composite laminates reinforced with glass and carbon Z-pins. Journal of Composite Materials 53(10): 1347–1359.

26.

Selver

Kaya

(2019b) Low velocity impact behaviour of carbon/XPS sandwich composites. Tekstil ve Mühendis 26(116): 353–359.

27.

Thomsen

Bozhevolnaya

Lyckegaard

(2005) Sandwich Structures 7: Advancing with Sandwich Structures and Materials. Aalborg: Springer.

28.

Virakthi

Kwon

Lee

(2018) A viable model for out-of-plane compressive and shear properties of Z-pin reinforced composite sandwich panels. Journal of Composite Materials 52(28): 3961–3972.

29.

Wallace

Sankar

Ifju

(2001) Pin reinforcement of delaminated sandwich beams under axial compression. Journal of Sandwich Structures and Materials 3(2): 117–129.

30.

Wang

, et al. (2010) Mechanical behavior of the sandwich structures with carbon fiber-reinforced pyramidal lattice truss core. Materials & Design 31(5): 2659–2663.

31.

Xiong

, et al. (2010) Fabrication and crushing behavior of low density carbon fiber composite pyramidal truss structures. Composite Structures 92(11): 2695–2702.

32.

Yalkin

Icten

Alpyildiz

(2015) Enhanced mechanical performance of foam core sandwich composites with through the thickness reinforced core. Composites Part B: Engineering 79: 383–391.

33.

Yang

Lei

, et al. (2020) Energy absorption and failure pattern of hybrid composite tubes under quasi-static axial compression. Composites Part B: Engineering 198: 108217.