Curved Polymeric Sandwich Composites Subjected To Air Shock
Curved Polymeric Sandwich Composites Subjected To Air Shock
https://doi.org/10.1007/s11340-024-01069-7
RESEARCH PAPER
Received: 30 January 2024 / Accepted: 8 April 2024 / Published online: 2 May 2024
© The Author(s) 2024
Abstract
Background The vulnerability of polymeric composite sandwich structures in marine applications to air explosions
highlights a significant gap in our understanding of the dynamic behavior of the curved sandwich structures, which is
essential for design improvements.
Objective This study aims to explore the dynamic response and failure mechanisms of curved sandwich composite panels
subjected to air-blast loading, providing insights into their structural integrity under such conditions.
Methods Experiments were performed using laboratory-simulated air shocks generated by a shock tube, employing high-speed
photography and digital image correlation to measure deflections on the back surface of the panels. The panels, made with PVC
closed-cell foam cores of two densities (H45 and H130), were tested across three curved geometries (radii of 112 mm, 305 mm,
and infinity) under various boundary conditions.
Results Findings indicate an increase in deformation with a decreased radius of curvature under simple support conditions, a
trend that reverses under arrested displacement conditions. Moreover, a reduced radius significantly enhances panel strength
and resistance to interfacial damage, with the primary failure mode transitioning from core shear cracking to interfacial
debonding as core density increases.
Conclusions The study reveals that the radius of curvature, boundary conditions, and core density significantly affect curved
sandwich panels’ dynamic response and performance. Panels with smaller radii and higher core densities exhibit increased
strength, though boundary conditions introduce variable effects on deformation behavior.
Keywords Sandwich composites · Curved structures · Blast loading · Shock loading · Progressive damage
Vol.:(0123456789)
946 Experimental Mechanics (2024) 64:945–961
loading conditions. Researchers have proposed passive tech- The results reveal that for a given impulsive loading on the
niques to improve blast performance, such as higher imped- curved sandwich panels with simple edge support at boundary,
ance mismatch at interfaces, graded core sandwiches, pro- deformation increases with the decrease in radius of curvature.
tective sacrificial cladding, laminate stacking with hybrid However, with arrested displacement at boundary, there is
interlaminar material configurations, and techniques for redi- reversal in the deformation trend. The sandwich panel with the
recting or disrupting the blast wave [2–6]. Significant con- least radius of curvature showed increase in the failure threshold
tributions have also been made to characterize the dynamic and capability to resist interfacial damage. The increase in the
response of polymeric sandwich composite structures to in- density of core material resulted in transition of primary failure
air blast loading through experiments. For instance, Wang mode from shear cracking in the core to interfacial debond.
et al. [7, 8] performed laboratory-scale air shock loading
experiments on sandwich beam specimens, revealing that
specific configurations of better-dissipated energy and in- Materials and Experimental Methods
plane compressive loads increase the damage to the face sheet
exposed to blast. Gardner et al. [9, 10] and Tekalur et al. [11] Specimen Design and Manufacturing
experimented with various sandwich panel configurations
to improve blast performance. Alanbay et al. [12] analyzed Specimen geometry selection
multilayered sandwich composite structures using a surro-
gate modeling-based optimization algorithm and developed The sandwich composite panels chosen for this study meas-
optimal core layup configurations for minimum deflections ured 200 mm wide, with a 2 mm thick facesheet and a 9.5
and transmitted force. mm thick core. The curvature radius of these panels was
Curved geometries are essential in naval structures due selected based on the principal area moment of inertia for
to design and manufacturing requirements, necessitating a curved cross-sectional sandwich. Equations (1) and (2)
understanding the geometric effects of curved sandwich were employed to evaluate this curved cross-section’s area
composite structures under shock loading. Previous stud- moment of inertia. Notably, the core’s contribution to the
ies, such as those by Kardomateas et al. [13], Rodcheuy area moment of inertia was disregarded since its modulus
et al. [14], Pradyumna et al. [15], and Hoo Fatt et al. [16], was considerably smaller than that of the facesheet material.
have developed various theories and models to predict the
dynamic behavior of curved sandwich structures. However, R44 − R43 + R42 − R41
Iyy = (2𝜃 − sin 2𝜃) (1)
limited experimental investigations exist on curved com- 8
posite panels exposed to blast loading. When loaded with
a shock tube, Kumar et al. [17] found that curved carbon R44 − R43 + R42 − R41
fiber-reinforced plastic laminates increased energy dissipa- Izz =
8
(2𝜃 + sin 2𝜃)
tion. Other researchers, like Langdon et al., Jing et al., and 4( 3
R4 − R33 + R32 − R31 y sin 𝜃 (2)
)( )
Shen et al. [18–20], reported similar findings on the dynamic −
3
behavior of curved sandwich panels under blast loading. Yet, )( 2 )
+ R4 − R23 + R22 − R21 y 𝜃
( 2
most of the literature on curved sandwich panels has focused
on postmortem observations.
where R1, R2 , R3, and R4 are the inner and outer radius of
This investigation captured the real-time blast response of
the bottom and top facesheet ( R1 < R2 < R3 < R4 ), y is
curved sandwich composite panels with different curvatures
the centroid distance of the curved cross-section sandwich
under blast-induced impulsive loading. The panels, consist-
composite from the center of curvature, and 2𝜃 is the angle
ing of a carbon fiber facesheet and a closed-cell PVC foam
subtended by the extreme edges at the center of the curved
core, were tested using a shock tube apparatus. The study
section. Figure 1(a) shows a schematic with the geometric
focuses on the following aspects –
details of a curved section.
The area moment of inertia ( Iyy and I zz) was evaluated
1. Quantification of impulse transferred due to planar shock
for the 178 mm wide curved cross-section sandwich with a
loading on single curve panels.
2 mm thick facesheet and 9.5 mm thick core. The variation
2. Influence of curvature and boundary conditions together
of the area moment of inertia with the inner radius of the
on the dynamic response.
curved section is shown in Fig. 1(b).
3. Role of core density on the failure of curved sandwich panels.
It was observed that when the inner radius is 600 mm or
4. Damage progression and postmortem observations in
greater, the area moment of inertia closely resembles that of
sandwich panels.
Experimental Mechanics (2024) 64:945–961 947
Y-axis
112 mm
Z-axis
305 mm
Centroid
R3
Flat panel
R1 R2 R4 305 mm
(a) (b)
Fig. 1 (a) Sketch of curved cross-section (b) Area moment of inertia variation with inner radius for curved cross-section
a rectangular cross-section, which is indicated by a dashed this research, labeled as ‘L1’ and ‘L2’, respectively. An
line in Fig. 1(b). However, the area moment of inertia exhib- in-depth explanation of these boundary conditions can be
its an exponential increase when the curved cross-section found in "Experimental Setup" section.
has an inner radius of less than 300 mm. For the scope of
this study, we chose two curved sandwich panels with inner Specimen manufacturing
radii of 112 mm and 305 mm, in addition to a flat sand-
wich panel. This selection encompasses a wide range of area The sandwich specimens were manufactured in-house using
moment of inertia values, as depicted in Fig. 1(b). the closed-cell PVC ( Divinycell® vinyl) foam provided by
In subsequent sections, the sandwich panels with inner DIAB International (Laholm, Sweden) as the core. Woven
radii of 112 mm, 305 mm, and infinity (representing the carbon fiber (balanced, twill) prepregs produced by GURIT
flat panel) are denoted as Panel A, Panel B, and Panel (Albacete, Spain) were utilized to fabricate the face sheets.
C, respectively. Figure 2 provides a schematic represen- The study used foam sheets with a thickness of 9.5 mm and
tation of the three chosen curved sandwich specimens, two distinct density grades: H45 and H130. An epoxy-based
showcasing their specific dimensions. This figure also resin film adhesive from GURIT was chosen to bond the
illustrates the distances (L) between the straight edges face-sheet to the core. These material details can be found in
for the two boundary conditions (BC1 and BC2) used in Table 1. The PVC foam core was thermoformed to achieve
Front facesheet
Fig. 2 (a) R112- Panel A (b) R305- Panel B (c) Flat- Panel C
948 Experimental Mechanics (2024) 64:945–961
Table 1 Raw material specification and properties The sandwich specimens were fabricated through the
Specification Density
vacuum bag molding process. This entailed co-curing the
woven prepreg with the thermoformed core, incorporating
Facesheet Twill woven prepreg 416 gsm [22] adhesive film at the interfaces. The curing was performed
(balanced)
at 85 °C for 9 hours, with a sustained vacuum pressure of 1
Core H45 52.1 ± 0.3 kg/m3
bar. Details of the manufacturing method, accompanied by
H130 143.9 ± 0.5 kg/m3
process parameters and images of representative sandwich
Film adhesive ST94 130 gsm
specimens, are presented in Fig. 3.
the desired curves. This process involved heating the foam Experimental Setup
sheets to a temperature of 5 °C below their maximum per-
missible processing temperatures [21] and pressing them The sandwich panels were subjected to air shock loading,
into a thermoforming mold for 30 min. and their dynamic response was captured on the back face
(d)
Fig. 3 Manufacturing process for sandwich composite panels (a) Thermoforming process (b) Curing cycle for sandwich panel (c) Schematic of
vacuum bag molding process (d) Manufactured specimens (typical)
Experimental Mechanics (2024) 64:945–961 949
sheet (opposite the loading face) using a high-speed stereo Boundary conditions and loading
DIC (Digital Image Correlation) method. Figure 4 illus-
trates a schematic of the experimental setup, highlighting The sandwich specimens, with their convex surfaces fac-
its essential components, such as the shock loading appa- ing the shock tube muzzle, were held in place using a
ratus, specimen boundary conditions, high-speed imaging fixture, as illustrated in Fig. 4. This study employed two
mechanisms, and the application of DIC. distinct fixtures to assess the impact of boundary condi-
tions. Figure 4(c) provides schematics of both fixtures,
Shock loading apparatus simulating two different boundary conditions: “Simple
edge support” and “Arrested displacements” at the straight
The study used a shock tube apparatus to produce controlled edges. For the simple edge support boundary condition,
planar shock waves for loading the specimens. This shock only the horizontal movement of the sandwich specimens
tube is primarily divided into a brief driver section and an was restrained. However, in the arrested displacements
extended driven section, possessing a circular cross-section. boundary condition, horizontal and vertical movements
The tube stretches over a length of 7 m. The inner diameters at the straight edges of the plates were inhibited, effec-
of the driver and driven sections measure 0.15 m, while the tively eliminating any motion in these directions. In the
muzzle section near the loading end has a diameter of 0.076 subsequent sections, the boundary conditions of simple
m. A stiff diaphragm membrane of mylar sheets separates edge support and arrested displacements are referred to
the driver and driven sections. Pressuring the driver section as ‘BC1’ and ‘BC2’, respectively.
until the membrane ruptures, the gas within quickly surges To assess the effect of boundary conditions on the panel’s
into the driven section. This abrupt motion generates a pres- response without causing damage under air shock loading,
sure pulse, culminating in a planar shock wave front. This specimens with the ‘BC1’ as well as ‘BC2’ boundary con-
shock wave then travels through the shock tube’s muzzle, ditions underwent a low-intensity shock named ‘LS’, which
making contact and applying pressure to the specimen. A had an approximate incident shock peak pressure magnitude
more detailed explanation of the operations of this apparatus of 0.5 MPa. Further, only the ‘BC1’ boundary condition was
can be found in previous work [23]. employed to investigate the effect of core density on the
950 Experimental Mechanics (2024) 64:945–961
shock response of the curved sandwich panels as well as the oscilloscope was channeled to the cameras, which prompted
damage progression in sandwich panels. The effect of core the commencement of high-speed image recording.
density on shock response was analyzed under ‘LS’ shock Table 2 provides a comprehensive overview of the experi-
intensity. And, more intense shock, labeled ‘HS’, which had ments presented in this study. The data encompasses both
an approximate incident shock peak pressure magnitude of the associated loads and the boundary conditions. Each spec-
0.8 MPa, was only used to induce damage in the sandwich imen is designated in the table using the format X-HYYY-
panels with high-density core. Compressed nitrogen gas was ZZ: ‘X’ signifies whether the panel is A, B, or C (as illus-
used in the driver section of the shock tube for the ‘LS’ trated in Fig. 3); ‘YYY’ represents the nominal core density
intensity loading. In contrast, a 70:30 pressure ratio blend in kg/m3; and ‘ZZ’ indicates the panel’s serial number. The
of Helium-Nitrogen gas facilitated the ‘HS’ shock loading. sandwich panels with low-density core were also subjected
to ‘LS’ shock load under boundary condition ‘BC2’. How-
Experimental Methodology ever, the obtained results were similar to the experimental
results presented here and, therefore, not included in this
The sandwich composite specimens, with their convex sur- manuscript to maintain brevity.
faces facing the shock tube muzzle, were secured using fix-
tures designed to simulate the boundary condition BC1 or
BC2. To generate the shock waves of the ‘LS’ intensity, the Results and Discussion
driver section of the shock tube was pressurized with Nitro-
gen gas. A Helium and Nitrogen gas mixture was utilized The results obtained from these experiments were compared
for the' HS' intensity shock. to understand:
Stereo Digital Image Correlation (DIC) was employed
to quantify the deformations on the back facesheet of the 1. The interplay between a planar shock front and a curved
sandwich panels, the side furthest from the loading end. structure.
Photron Fastcam Nova S12 cameras equipped with a 60 2. The influence of curvature on structural response under
mm AF micro Nikkor lens (from Nikon, Japan) were used varied boundary conditions.
to acquire high-speed images. These cameras captured 3. The impact of core density on the structural response.
images at 25,000 frames per second with a pixel resolution 4. Damage progression and postmortem observations in
of 768 × 640. A Photron Fastcam SA1.1 camera featuring sandwich panels.
a 105 mm AF-DC Nikkor lens (also from Nikon, Japan)
was also positioned perpendicular to the shock tube’s axis. Interaction of Shock Wave with Curved Geometries
Its purpose was to record any deformation and damage
within the core. This camera operated at 20,000 frames In the shock tube, both the incident and reflected shock pres-
per second and provided a pixel resolution of 192 × 832. sures acting on the sandwich specimen were measured using
The DIC analysis of the panel deformation was performed the pressure sensor ‘S1’ positioned on the shock tube muz-
using a commercial software package: VIC 3D ver.9, zle [23]. Under normal incidence, the reflected shock pres-
developed by Correlated Solutions, located in Columbia, sure,P0 , is influenced solely by the incident shock strength
South Carolina. (represented as y = Pi ∕Patm )) and the ratio of the specific
Dynamic pressure sensors (Models 113B22 & 102B04) heat constants (given as𝛾 = Cp ∕Cv ), as shown in equations
from PCB Piezotronics were employed to measure the shock (3) and (4).
pressure. Pressure data acquisition was recorded using a
Tektronix digital oscilloscope (Model: DPO 3034), operat- P0 (2𝜇 + 1)y − 𝜇
ing at a 25 MS/s sampling rate. The trigger output from the
= (3)
Pi 𝜇y + 1
Table 2 List of experiments Exp. No. Specimen Load BC Exp. No. Specimen Load BC
0.5 0.5
Incident shock Incident
0 0 shock
-0.5 -0.5
0 5 10 15 20 25 0 5 10 15 20 25
Time [in ms] Time [in ms]
(a) (b)
Fig. 5 Typical pressure–time history for shock loading (a) Low intensity (b) High intensity
Fig. 6 Reflected shock pressure variation along the cylindrical surface (a) Pressure ratio variation with angle 𝜽 (b) Planar shock interaction with
cylindrical structure (Schematic) [25] (c) Curved panel exposed to a planar shock through a circular cross-section muzzle (Schematic)
√
∫
𝛼
∫
r2 (6)
As = 4R2 − sin2 𝜃 d𝜃 (5) F(t) = ((P0 (t)cos 𝜃 + Pi (t)) cos 𝜃)dA
0 R2 A
∫
exposed to shock was calculated by considering ‘ r ’ as 38.1 r2
2
F(t) = 𝜋r Pi (t) + 4Po (t)R 2 2
cos 𝜃 − sin2 𝜃 d𝜃 (7)
mm, and ‘ R ’ as 125.5 mm, and 318.5 mm for Panels A and 0 R2
B, respectively. The net pressure distribution at a given Equation (8) gives the impulse I , transferred to the
instant ‘t’ on the cylindrical surface can be mathematically cylindrical structure during shock loading, which is the
written as P(t) = P𝜃 (t) + Pi (t). Due to the symmetry of the time integral of the instantaneous force F(t) . The impulse
curved sandwich panel, the net force acting on the struc- is expressed in terms of the specific impulse (N-s/m2),
ture is only due to the cosine component of the normal due to the incident pressure, I0inc = ∫ 0 Pi (t)dt and reflected
T+
force, i.e., the force component along the direction of the pressure, I0 = ∫ 0 Po (t)dt under the normal incidence of
ref T+
traveling shock, and the net force on a panel at any given shock, for the duration of over-pressure.
instant ‘t’ can be mathematically represented by equation
(6). Further substituting the results in the expression for 𝛼
√
∫ 0
r2
instantaneous force, F(t) as shown in equation (7).
2 inc
I = 𝜋r I0 + 4R Io2 ref 2
cos 𝜃 − sin2 𝜃 d𝜃 (8)
R2
Experimental Mechanics (2024) 64:945–961 953
Table 3 Impulse due to planar shock from the shock tube shockwave, had a marginal increase of less than 1% in
Sandwich Panel Impulse (N-s) Normalized
impulse compared to Panel A.
impulse
Fig. 7 DIC results for low- Panel A Panel B Panel C (in mm)
intensity shock loading on 10
curved sandwich panels with 15
high-density PVC closed-cell Panel A
Panel B
foam core (a) Out-of-plane Displacement (mm) Panel C
10
2.5
5
0
0 1 2 3 4 5
Time (ms)
0 (c)
(a)
954 Experimental Mechanics (2024) 64:945–961
the stereo DIC camera setup, as detailed in "Experimental condition. Also, for the ‘BC2’ boundary condition, Panel A
Setup" section. Figure 7(a) presents typical deformation pro- displayed the least peak deflection compared to the other two
files of sandwich panels with varying curvatures. panels. This starkly contrasts the behavior noted for ‘BC1’
Figure 7(b) shows that for the first 0.5 ms, Panel A dis- boundary conditions, where Panel A demonstrated the maxi-
played the least center point displacement among all the mum peak deflection. Figure 7(c) shows a contrasting pattern
sandwich panels with the ‘BC1’ boundary condition. The in maximum deflection at the center of the specimens with
center point displacement was more pronounced for the boundary condition ‘BC2’. This opposite trend in peak deflec-
panel with a larger curvature radius. However, even with tions is the influence of the boundary conditions. This change
the slowest center displacement rate, Panel A displayed the in peak deflection behavior is due to the nonlinear distribu-
largest initial peak of center point displacement compared to tion of bending moment along the curved surface, which is
the other panels. Figure 7(c) shows the center point deflec- more pronounced for the panels with greater angular extent.
tion of the sandwich panels following the ‘BC2’ boundary The angular extent is the angle subtended by the boundary
condition. These sandwich panels registered a deflection supports at the panel’s center of curvature. Xie et al. [26]
magnitude notably lower than the peak deflections observed observed similar behavior in their quasi-static three-point
in panels governed by the ‘BC1’ boundary. bend tests on curved sandwich beams.
Displacement (mm)
15
15 10
0
0 0.3 0.6 0.9 1.2
Time (ms)
10 (b)
25
Panel A
Panel B
20 Panel C
Displacement (mm)
15
5 10
0
0 1 2 3 4 5
Time (ms)
(a) 0 (c)
Fig. 8 DIC results for curved sandwich panels with BC1 boundary condition exhibiting damage under shock loading (a) Out-of-plane displace-
ment contours for high-density core sandwich panels (Typical) (b) Center point out-of-plane displacements for low-density (H45) core sandwich
panels under low-intensity ‘LS’ shock loads (c) Center point out-of-plane displacements for high-density (H130) core sandwich panels under
high-intensity ‘HS’ shock loads
Experimental Mechanics (2024) 64:945–961 955
Sandwich panels with low‑density (H45) core Failure in Highest Curvature Sandwich Panel The first
visible damage in Panel A was observed as a shear crack
All three sandwich panels, A, B, and C, embedded with low- within the core at 900 μs. This crack advanced toward the
density core, showed catastrophic failure at ‘LS’ intensity front and back facesheets and transformed into an interfa-
shock loading. Also, panels B and C demonstrated analo- cial crack upon reaching the facesheet-core interface. As
gous damage progressions with this low-density core. Thus, Fig. 9(a) shows, crack propagation eventually led to interfa-
their response is collectively shown in Fig. 9. Only one of cial de-bonding between the core and facesheet. The sepa-
these panels (Panel B) is compared with panel A to highlight rated front facesheet, having lost its stiffness, underwent
the similarities and differences in the failure modes for these significant bending, and failed under compressive stress
curved composite sandwich panels. at 3500 μs.
Fig. 10 Postmortem observations on sandwich panels with low-density closed-cell PVC foam core (a) Panel A, (b) Panel B, (c) Panel C (d)
Schematic of damage propagation in low-density core sandwich panels
Experimental Mechanics (2024) 64:945–961 957
Failure in Intermediate Curvature and Flat Sandwich Pan‑ of sandwich Panel A highlighted shear cracking along the
els The onset of damage in Panel B was marked by a crack panel’s width. These shear cracks, inclined to the panel’s
at 600 μs as seen from the side view camera. This crack was curved profile, were also recorded through high-speed pho-
positioned proximate to the back facesheet of the sandwich tography, as presented in Fig. 9(a). Panel A did not exhibit
panel. As it expanded, it vertically split the panel into two core crushing, unlike Panels B and C, which demonstrated
sections. A similar pattern of core damage was observed in evident core crushing during post-test analysis. The core of
Panel C. Overall, Panels B and C exhibited damage earlier sandwich Panel B displayed a unique elliptical failure pat-
and more rapidly than Panel A. tern, as depicted in Fig. 10(b). The back facesheet retained an
elliptically shaped core segment, creating a matching cavity
Figure 10 presents the images of the damaged sandwich in the front facesheet exposed to shock loading. A minor
panels comprising a low-density core. A post-examination foam fragment was also observed at the center of the front
Core
crack
Curved Panel
(Panel B) Interface
de-bond
Facesheet
Buckling
(a)
Flat Panel
(Panel C)
(b)
958 Experimental Mechanics (2024) 64:945–961
facesheet. The flat sandwich panel, Panel C, broke into two • Stage 4: The aftermath of a failed sandwich beam manifests
distinct sections. The core largely remained attached to the as a void in the front facesheet, complemented by a corre-
back facesheet, leaving traces in the central section and cor- sponding section of the core attached to the back facesheet.
ners of the front facesheet, as illustrated in Fig. 10(c). An
elliptical failure contour characterized Panel B’s core, while In these experiments, the transverse shock loading on
a circular shape was evident in Panel C’s core. Figure 10(d) the wide sandwich panels manifests as the failure surface
offers a schematic elucidating this elliptical/circular core adopting a round geometric shape. The failure contours for
pattern through a step-by-step damage progression in the Panel B and Panel C shown in Fig. 10(b) and (c) are ellip-
sandwich beam, delineating the damage evolution into four tical and circular geometry. The ellipticity of this contour
distinct phases: changes with the curvature-induced stiffness inherent to
the sandwich panel.
• Stage 1: The shockwave emanating from the shock tube Furthermore, based on the high-speed imaging and post-
muzzle induces indentation in the low-density core, caus- mortem observations of sandwich panels (Panel B and Panel
ing core crushing. Concurrently, specimen deformation C) with a low-density core subjected to 'LS' shock loading
occurs, leading to shear cracks within the core. under 'BC2' boundary conditions, no discernible effect of
• Stage 2: These shear cracks extend towards the boundary conditions was observed on the progression of
facesheets, culminating in interfacial cracks. damage or failure contours.
• Stage 3: These interfacial cracks prompt the interfacial Besides the direct shock effects, damage to the sandwich
de-bonding between the core and the facesheet over time. panels also stems from the propagation of stress waves through
Delamination in
Front facesheet
Flexure cracks
in core
Delamination in
Back facesheet Crack plane normal to curved profile
(a)
Back facesheet
(b)
Fig. 12 Postmortem observations on sandwich panels with high-density closed-cell PVC foam core (a) Panel B (b) Panel C
Experimental Mechanics (2024) 64:945–961 959
the panel. Damage in the sandwich core was estimated by interface occurred, a phenomenon captured at 2250 μs, as
assuming the linear elastic stress wave propagation through shown in Fig. 11(a).
the facesheet and core materials. Analytical results revealed Sandwich Panel C, in its post-experiment state shown in
that the stress wave amplitude in the core stands at 90 kPa in Fig. 12(b), exhibited a single flexure crack located centrally.
compression, a value substantially below the core’s crushing Interfacial de-bond was unmistakably evident, given the adhe-
strength (600 kPa at a strain rate of 103/sec, as reported by sive remnants on both the core and the face sheet, further high-
Wanchoo et al. [27]). Thus, the damage contributed to the core lighted in Fig. 12(b). Another noteworthy observation was the
is primarily due to the deformation in sandwich panels. delamination within the face sheets of both Panels B and C.
foam. However, the boundary conditions do not affect 4. Langdon GS, Cantwel WJ, Guan ZW, Nurick GN (2014) The response
the damage progression and failure surface. The sand- of polymeric composite structures to air-blast loading: a state-of-the-
art. 59:159–77. https://doi.org/10.1179/1743280413Y.0000000028
wich panels with low-density cores primarily fail from 5. Wanchoo P, Matos H, Rousseau CE, Shukla A (2021) Investiga-
core cracking due to transverse shear stresses. However, tions on air and underwater blast mitigation in polymeric compos-
this failure mode changes to interfacial de-bond at the ite structures – A review. Compos Struct 263:113530. https://doi.
core-facesheet interface towards the shock. org/10.1016/J.COMPSTRUCT.2020.113530
6. Wanchoo P, Pandey A, Leger M, LeBlanc J, Shuka A (2024)
Energy quantification framework for underwater explosive load-
ing into PVC foam cladded composite plates. J Mech Phys Sol-
ids. https://doi.org/10.1016/j.jmps.2024.105646
Funding The authors gratefully acknowledge the Office of Naval 7. Wang E, Gardner N, Shukla A (2009) The blast resistance of sand-
Research (ONR) for the financial support provided under Grant No. wich composites with stepwise graded cores. Int J Solids Struct
N00014-20–1-2877. The authors would also like to acknowledge the 46:3492–3502. https://doi.org/10.1016/j.ijsolstr.2009.06.004
DIAB group for providing the PVC foam materials used in the experi- 8. Wang E, Shukla A (2012) Blast performance of sandwich com-
mental work. The authors are also grateful to their colleagues in the posites with in-plane compressive loading. Exp Mech 52:49–58.
Dynamic Photo-mechanics Laboratory at URI for their support during https://doi.org/10.1007/s11340-011-9500-5
the experimentation, especially Chris Grivers, Jacob Milman, Jamie 9. Gardner N, Wang E, Shukla A (2011) Performance of functionally
Davis, Keevan Winters, and Tyler Chu. graded sandwich composite beams under shock wave loading.
https://doi.org/10.1016/j.compstruct.2011.12.006
Data Availability The data used in this study will be made available by 10. Gardner N, Wang E, Kumar P et al (2012) Blast mitiga-
the corresponding author upon reasonable request. tion in a sandwich composite using graded core and polyurea
interlayer. Exp Mech 52:119–133. https:// d oi. o rg/ 1 0. 1 007/
Declarations s11340-011-9517-9
11. Tekalur SA, Bogdanovich AE, Shukla A (2009) Shock loading
Conflict of Interest The authors certify that they have no affiliations response of sandwich panels with 3-D woven E-glass composite
with or involvement in any organization or entity with any financial in- skins and stitched foam core. Compos Sci Technol 69:736–753.
terest (such as honoraria; educational grants; participation in speakers’ https://doi.org/10.1016/J.COMPSCITECH.2008.03.017
bureaus; membership, employment, consultancies, stock ownership, 12. Alanbay B, Batra RC (2022) Optimization of blast mitigating
or other equity interest; and expert testimony or patent-licensing ar- sandwich structures with fiber-reinforced face sheets and PVC
rangements), or non-financial interest (such as personal or professional foam layers as core. Thin-Walled Struct 179:109721. https://doi.
relationships, affiliations, knowledge or beliefs) in the subject matter or org/10.1016/J.TWS.2022.109721
materials discussed in this manuscript. 13. Kardomateas GA, Rodcheuy N, Frostig Y (2017) Elasticity solu-
tion for curved sandwich beams/panels and comparison with
Open Access This article is licensed under a Creative Commons Attri- structural theories. AIAA J 55:3153–3160. https://doi.org/10.
bution 4.0 International License, which permits use, sharing, adapta- 2514/1.J055760
tion, distribution and reproduction in any medium or format, as long 14. Rodcheuy N, Frostig Y, Kardomateas GA (2017) Extended high-
as you give appropriate credit to the original author(s) and the source, order theory for curved sandwich panels and comparison with
provide a link to the Creative Commons licence, and indicate if changes elasticity. J Appl Mech Trans ASME. https://doi.org/10.1115/1.
were made. The images or other third party material in this article are 4036612/474273
included in the article’s Creative Commons licence, unless indicated 15. Pradyumna S, Nanda N, Bandyopadhyay JN (2010) Geometrically
otherwise in a credit line to the material. If material is not included in nonlinear transient analysis of functionally graded shell panels
the article’s Creative Commons licence and your intended use is not using a higher-order finite element formulation. J Mech Eng Res
permitted by statutory regulation or exceeds the permitted use, you will 2:39–51
need to obtain permission directly from the copyright holder. To view a 16. Fatt MSH, Gao Y, Sirivolu D (2013) Foam-core, curved composite
copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. sandwich panels under blast. J Sandwich Struct Mater 15:261–
291. https://doi.org/10.1177/1099636213481469
17. Kumar P, Stargel DS, Shukla A (2013) Effect of plate curvature
on blast response of carbon composite panels. Compos Struct
References 99:19–30. https://d oi.o rg/1 0.1 016/J.C
OMPST RUCT.2 012.1 1.0 36
18. Langdon GS, Von Klemperer CJ, Sinclair GM (2016) Blast
1. Mouritz AP, Gellert E, Burchill P, Challis K (2001) Review of response of sandwich structures: the influence of curvature.
advanced composite structures for naval ships and submarines. https://doi.org/10.1016/B978-0-08-100870-6.00013-4
Compos Struct 53:21–42. https://d oi.o rg/1 0.1 016/S
0263-8 223(00) 19. Jing L, Wang Z, Shim VPW, Zhao L (2014) An experimental
00175-6 study of the dynamic response of cylindrical sandwich shells with
2. Wanchoo P, Kishore S, Pandey A, Shukla A (2022) Advances in metallic foam cores subjected to blast loading. Int J Impact Eng
naval composite and sandwich structures for blast and implosion 71:60–72. https://doi.org/10.1016/J.IJIMPENG.2014.03.009
mitigations. J Sandwich Struct Mater. https://doi.org/10.1177/ 20. Shen J, Lu G, Wang Z, Zhao L (2010) Experiments on curved
10996362221139562 sandwich panels under blast loading. Int J Impact Eng 37:960–
3. Birman V, Kardomateas GA (2018) Review of current trends in research 970. https://doi.org/10.1016/j.ijimpeng.2010.03.002
and applications of sandwich structures. Compos B Eng 142:221–240. 21. MECHANICAL PROPERTIES DIVINYCELL ® H 2023. https://
https://doi.org/10.1016/J.COMPOSITESB.2018.01.027 diab-media.azureedge.net/eyajkrhd/diab-divinycell-h-october-
2023-rev23-si.pdf
Experimental Mechanics (2024) 64:945–961 961
22. Full General Datasheet ST 130FR 130 °C TG FIRE RETARD- 27. Wanchoo P, Pandey A, Shukla A (2023) Stress wave propagation
ANT LOW SMOKE SPRINT TM. https://www.gurit.com/wp- and force transmission in polymeric closed cell foams subjected
content/uploads/bsk-pdf-manager/2022/08/ST-130FR.pdf to air shock loading. Extreme Mech Lett 63:102061. https://doi.
23. Wang E, Shukla A (2010) Analytical and experimental evalu- org/10.1016/J.EML.2023.102061
ation of energies during shock wave loading. Int J Impact Eng 28. Thomas T, Mahfuz H, Carlsson LA, Kanny K, Jeelani S (2002)
37:1188–1196. https://doi.org/1 0.1016/J.IJIMPENG.2 010.0 7.003 Dynamic compression of cellular cores: temperature and strain
24. Wanchoo P, Chaudhary B, Li HWX, Matos H, Shukla A (2023) rate effects. Compos Struct. https://d oi.o rg/1 0.1 016/S 0263-
Blast failure and energy analysis of rubber-modified carbon-fiber 8223(02)00159-9
vinyl-ester composite laminates. Mech Mater 183:104685. https://
doi.org/10.1016/J.MECHMAT.2023.104685 Publisher's Note Springer Nature remains neutral with regard to
25. Glasstone S, Dolan PJ (1977) The effects of nuclear weapons. jurisdictional claims in published maps and institutional affiliations.
Third edition. https://doi.org/10.2172/6852629
26. Xie H, Li W, Fang H, Zhang S, Yang Z, Fang Y et al (2024)
Flexural behavior evaluation of a foam core curved sandwich
beam. Compos Struct 328:117729. https://doi.org/10.1016/J.
COMPSTRUCT.2023.117729