FIBER REINFORCED HIGH STRENGTH CONCRETE WITH CELLULAR STEEL SANDWICH COMPOSITE PANEL FOR BLAST AND IMPACT MITIGATION

FIBRE REINFORCED HIGH STRENGTH CONCRETE WITH CELLULAR STEEL SANDWICH COMPOSITE PANEL FOR BLAST AND IMPACT MITIGATION ABRAHAM CHRISTIAN NATIONAL UNIVERSITY OF SINGAPORE 2015... FIBRE RE

FIBRE REINFORCED HIGH STRENGTH CONCRETE WITH CELLULAR STEEL SANDWICH COMPOSITE PANEL FOR BLAST AND IMPACT MITIGATION

ABRAHAM CHRISTIAN

NATIONAL UNIVERSITY OF SINGAPORE

2015

FIBRE REINFORCED HIGH STRENGTH CONCRETE WITH CELLULAR STEEL SANDWICH COMPOSITE PANEL FOR BLAST AND IMPACT MITIGATION

ABRAHAM CHRISTIAN

(B.Eng.(Hons.), Diponegoro University)

A THESIS SUBMITTED FOR THE DEGREE OF DOCTOR OF PHILOSOPHY DEPARTMENT OF CIVIL AND ENVIRONMENTAL ENGINEERING

NATIONAL UNIVERSITY OF SINGAPORE

2015

Declaration Page

DECLARATION

I hereby declare that this thesis is my original work, and I have written it in its entirety

I have duly acknowledged all the sources of information which have been used in the thesis This thesis has also not been submitted for any degree in any university previously

Abraham Christian February 2015

Acknowledgements

All praise to God, only by His grace this study can be completed

The author wishes to express his sincere gratitude to his supervisors, Assoc Prof Ong Khim Chye, Gary for his personal commitment, patience, interesting discussion, invaluable guidance and constructive advices throughout the course of this study

The author’s heartfelt appreciation is dedicated to Dr Lado R Chandra, Dr Patria Kusumaningrum and Dr Satadru Das Adhikary for their contributions and continuous supports

Sincere thanks are also extended to the DSTA for providing the funding required for blast experimental testing

The kind assistance from all the staff members of the NUS Concrete and Structural Engineering Laboratory is deeply appreciated Special thanks go to Mr Koh, Mr Ang, Mr Ishak, Mr Stanley and Mr Kamsan for their continuous support during experimental phase

of the study

Finally, special thanks and loves go to my parents, brother, sister and friends for the moral supports and constant love Thank you for making this study possible and may God bless all of you

Table of Contents

Declaration Page i

Acknowledgements ii

Table of Contents iii

Summary viii

List of Tables x

List of Figures xi

List of Symbols xviii

Chapter 1 Introduction 1

Background of the study 1

Literature review 7

Composite panel as an energy absorber 7

Metal sandwich panels 7

Steel-concrete-steel (SCS) sandwich panels 12

Fibre composite panels 14

Steel stud composite panel 18

Multi-material layered composite panel 19

Observations arising from the literature review 20

Objectives and scope of study 22

Thesis structure 24

Chapter 2 Proposed Composite Panel 26

Design criteria 26

Proposed composite panel design 29

Constituent composite sections 30

Steel sandwich core structure design 32

Steel sandwich fabrication techniques 35

Connection details for handling and installation 37

Finalized panel design 37

RC specimen fabrication for performance comparison 38

Material usage and cost comparison 39

Fabrication process 40

Chapter 3 Finite Element Analysis 43

Introduction 43

General parameters of EASP FE model 44

Mesh convergence study 44

Development of EASP model 45

Material Models 47

Strain rate effect 59

Contact model 63

Hourglass 64

Bulk viscosity for dynamic loading case 65

Erosion criteria for solid elements 66

Other LS-Dyna keyword 66

Validation of FE model 66

Numerical study on the EASP energy absorption 68

Simulation of three point static bending of EASP and RC panels 72

Simulation of EASP specimens subjected to impact loading 74

Simulation of EASP specimens subjected to blast loading 78

Blast wave loading from cylindrical charges 79

Summary of finite element analyse of EASP specimens 83

Chapter 4 Static Performance of EASP 84

Introduction 84

Composite panel under static loading 84

Specimen details 85

RC specimen fabrication for performance comparison 85

Moment capacity analysis 86

Testing setup and instrumentation 92

Results and discussion 93

Load deflection response 93

Failure mode and cracking behaviour of the concrete layer 95

Steel plate strain data 98

Effect of concrete strength and steel fibre incorporation 100

Comparison to analytical and numerical result 101

Conclusions 108

Chapter 5 Impact Performance of EASP 110

Introduction 110

Impact studies on reinforced concrete and steel-concrete composite panel 111

Failure modes 113

Testing method 115

Specimens 115

Test set-up 117

Results and discussion 119

Crack propagation and final crack pattern 120

Damage in the concrete and steel-sandwich layers 127

Displacement-time history 132

Strain measurement on the steel sandwich interface and distal plates 138

Comparison with numerical results 142

EASP with full core configuration 142

EASP with hollow core configuration 145

Discussion 149

EASP with various concrete materials 149

EASP full core versus hollow core 150

EASP versus RC 151

Conclusions 151

Chapter 6 EASP Blast Resistance 153

Introduction 153

Air blast wave in explosions 153

Composite structure subjected to blast loading 160

Testing method 163

Specimens 163

Experimental set-up 164

Results and discussion 165

Pressure histories 168

Displacement-time history 174

Damage assessment 175

Conclusions 177

Chapter 7 Conclusions and Recommendations 179

Review on completed research 179

Conclusion summary 179

Recommendations for future studies 183

References 185

Appendices 190

Publications 191

Summary

The study investigated the performance of concrete-steel composite panels subjected to static and dynamic loading The panel consists of fibre-reinforced high strength concrete overlaid on top of a specially configured steel sandwich, which function to dissipate the imparted kinetic energy The core structure of the steel sandwich resists mostly tensile forces while absorbing the energy through plastic buckling The novel composite panel was assigned the name Energy Absorption Sandwich Panel or EASP The intention was to use it

as a secondary protection or sacrificial cladding panel in existing buildings to shield critical structural components against blast

Experimental testing was done in three phases; three point static bending tests, velocity drop-weight impact tests and close-in cylinder TNT explosion tests Parametric studies were carried out utilizing two types of steel sandwich core structure (bended and straight type) as well as various concrete materials: normal strength concrete (NSC) 60 MPa, high strength concrete (HSC) 110 MPa and fibre reinforced high strength concrete (FRHSC) 110 MPa Ordinary reinforced concrete panels of the same geometric dimensions and tension capacity cast with HSC and FRHSC were also included for benchmarking The

low-RC panel was on average about 33% heavier than the EASP

The response of the EASP against static bending, impact and blast loading was obtained through experimental tests The performance assessed by observing the damage to the constituent concrete and steel sandwich layers, comparing the maximum and residual deformation and observing the failure modes of the different types of panels It was found that under static bending tests, the maximum flexural resistance did not differ much although the EASP could achieve much higher ductility vis-à-vis the high strength

reinforced concrete panel Under the drop weight impact tests, the EASP exhibited

improved resistance with reduced damage compared with the RC panels The better

performance of EASP with FRHSC concrete material is characterized by less residual and

maximum deflection, reduced concrete fragmentation and large reduction in concrete

crack propagation The EASP in general cast using various concrete types and sandwich

cores were able to withstand 800kg drop weight impact loading at an impact velocity of

6.26 m/s The similar high strength reinforced concrete panel failed with full projectile

penetration These demonstrated that EASP performed better and possess higher energy

absorption capacity compared with the RC panel Last but not least, when subjected to

close-in TNT blast loading, the EASP could withstand blast pressures of up to 50 kg of TNT

at 1m standoff distance (equalized mass from cylinder to spherical charges) with minimum

concrete spalling It showed that the proposed panel has potential to be used as sacrificial

cladding protective panel to resist close-in blast loading

Finite element simulation using the LS-Dyna software was used to model various types of

panel when subjected to loading used in the three phase tests The simulation results could

predict the panels’ response against static and dynamic loading with good accuracy The

combination of fibre reinforced high strength concrete and steel sandwich structure

demonstrated good potential for use as blast and impact mitigation protective panels with

better weight-performance ratios and enhanced energy absorption properties

Keywords: steel-concrete composite, drop-weight impact, blast loading, fibre reinforced

high strength concrete, steel sandwich structure, energy absorption

List of Tables

Table 1-1 List of bombing incidents (wikipedia.org, 2012) 1

Table 3-1 Details of model simulation for mesh convergence study 45

Table 3-2 Concrete mix design 52

Table 3-3 Concrete fracture energy for various strengths and aggregate size (CEB-FIP 1990)………… 55

Table 3-4 Concrete material properties 57

Table 3-5 Steel material properties 58

Table 3-6 Steel rebar & steel plate material properties 59

Table 4-1 Static test specimen details 85

Table 4-2 Bending resistance calculation table (with m) 89

Table 4-3 Bending resistance calculation table (without m) 90

Table 4-4 Comparison of analytical, numerical and experimental results 108

Table 5-1 Impact test specimens & loading details 117

Table 6-1 Explosives conversion factor 156

List of Figures

Figure 1-1 Catastrophic failure of Alfred P Murrah building 3

Figure 1-2 The aftermath of Khobar housing complex bombing 4

Figure 1-3 Transmission of blast load from the secondary barrier 6

Figure 1-4 Sandwich panel typical geometry 8

Figure 1-5 Metal sandwich structures (CEL-Components, 2012) 8

Figure 1-6 Deformed shape for steel sandwich (Guruprasad & Mukherjee, 2000b) 9

Figure 1-7 Alporas aluminium foam structure (AlCarbon Technology company) 10

Figure 1-8 Maximum blast impulse sustained by monolithic beams (Fleck & Deshpande, 2004)……… 11

Figure 1-9 Bi-Steel SCS panel (TATA-STEEL, 2012) 13

Figure 1-10 Bridged crack in FRHSC (BrightHub-Eng., 2011) 15

Figure 1-11 Steel stud wall construction (before and after blast) 18

Figure 2-1 Early Design of Energy Absorption Sandwich Panel (EASP) 29

Figure 2-2 Basic principle of the proposed panel 30

Figure 2-3 (a) Typical Structure of SCS; (b) Cellular Sandwich Panel 33

Figure 2-4 Six Types of Steel Sandwich Core Configuration 33

Figure 2-5 Groups of nodes desgnated as fixed support 34

Figure 2-6 Distal Plate Maximum Displacement 34

Figure 2-7 Distal Plate Springback Displacement 35

Figure 2-8 Slot joint design 36

Figure 2-9 Panel connection to supporting structural element 37

Figure 2-10 3D Assembly of easp1 38

Figure 2-11 Rebar details for RC specimen 39

Figure 2-12 Steel Sandwich layer with reinforcing bars 41

Figure 2-13 EASP1 ready for concrete casting (left) and soffit of EASP2 (right) 41

Figure 2-14 Detailed dimensions of EASP1 & EASP2 42

Figure 2-15 CrossS Section of EASP1 & EASP2 42

Figure 3-1 FE model of EASP1 and EASP2 (quarter model) 46

Figure 3-2 Model of alternating welding joint in steel sandwich layer 47

Figure 3-3 (a) Concrete model failure surface and (b) Material model stress-strain curve……… 48

Figure 3-4 Concrete mean strength (fcm) 52

Figure 3-5 Concrete characteristic strength (fck) 53

Figure 3-6 Single element uniaxial compressive test 53

Figure 3-7 Stress-strain curve of C60 concrete element model 54

Figure 3-8 Stress-strain curve of C110 concrete element model 54

Figure 3-9 Stress-strain curve of C110F concrete element model 55

Figure 3-10 C60 and C110 concrete model tensile stress-strain curve 56

Figure 3-11 Steel coupon stress-strain curve 58

Figure 3-12 Elastic-plastic steel behaviour with kinematic and isotropic hardening 59

Figure 3-13 Stress-strain curves of concrete at different strain rates (T Ngo, Mendis, P., Hongwei, M & Mak, S., 2004) 60

Figure 3-14 Load-displacement history of RC-C110 specimen in comparison with numerical prediction…… 67

Figure 3-15 Load-displacement history of RC-C110F specimen in comparison with numerical prediction…… 67

Figure 3-16 Displacement-time history of FRHSC with 10kg; 0.5m SoD blast 69

Figure 3-17 Maximum displacement values of all the panels subjected to various blast

charges: (a) FRHSC; (b) HSC 70

Figure 3-18 Resistance-deflection curve at 10kg, 0.5m SoD charge of FRHSC materials 71

Figure 3-19 EASP1 quarter model for static bending simulation 73

Figure 3-20 Numerical load-displacement history EASP1 specimens 73

Figure 3-21 Numerical load-displacement history EASP2 specimens 74

Figure 3-22 Quarter model of the EASP1 specimen for impact simulation 76

Figure 3-23 EASP1 specimens FE simulation results 76

Figure 3-24 EASP2 specimens FE simulation results 77

Figure 3-25 EASP1H specimens FE simulation results 77

Figure 3-26 EASP2H specimens FE simulation results 78

Figure 3-27 Blast pressure prediction for perpendicular charge 80

Figure 3-28 Zone division for applying approximated blast pressure 80

Figure 3-29 EASP1-C110F specimen blast simulation results 81

Figure 3-30 Blast loading approximation for parallel charge 82

Figure 3-31 EASP1-C110 specimen blast simulation results, a comparison 82

Figure 4-1 Rebar details for RC specimen 86

Figure 4-2 Rectangular concrete stress block 86

Figure 4-3 Strain-stress progression from SLS to ULS 87

Figure 4-4 RC compression-tension forces 87

Figure 4-5 EASP1 and EASP2 compression-tension forces 88

Figure 4-6 Moment-curvature curve of EASP2-C60 90

Figure 4-7 Moment-curvature curve of EASP2-C110 91

Figure 4-8 Moment-curvature curve of EASP2-C110F 91

Figure 4-9 Static test setup diagram for EASP specimens 92

Figure 4-10 EASP1 & RC load-displacement history 93

Figure 4-11 EASP2 & RC load-displacement history 95

Figure 4-12 EASP1 and 2 (C60) specimens crack pattern at Ultimate Load 97

Figure 4-13 EASP1 and 2 (C110) specimens crack pattern at Ultimate Load 97

Figure 4-14 EASP1 and 2 (C110F) specimens crack pattern at Ultimate Load 97

Figure 4-15 RC-C110 and RC-C110F specimens crack pattern at Ultimate Load 97

Figure 4-16 Steel plate strain gauges data 99

Figure 4-17 Crack pattern of (a) RC-C110F and (b) EASP2-C110F 100

Figure 4-18 Load-displacement history and cracking pattern of EASP1-C60 specimen 102

Figure 4-19 Load-displacement history and cracking pattern of EASP1-C110 specimen 102 Figure 4-20 Load-displacement history and cracking pattern of EASP1-C110F specimen 103 Figure 4-21 Load-displacement history and cracking pattern of EASP2-C60 specimen 103

Figure 4-22 Load-displacement history and cracking pattern of EASP2-C110 specimen 104 Figure 4-23 Load-displacement history and cracking pattern of EASP2-C110F specimen 105 Figure 4-24 Load-displacement history of RC-C110 specimen 106

Figure 4-25 Load-displacement history of RC-C110F specimen 106

Figure 4-26 Crack propagation of RC-C110F specimen 107

Figure 5-1 J-Hook SCS Sandwich Panel (Liew et al., 2009) 112

Figure 5-2 Deflection profile of various projectile loading rates 113

Figure 5-3 Full core EASP1 (a) and hollow core configuration EASP1H (b) 116

Figure 5-4 Hollow core area of EASP1H (B) compared to EASP1 (a) 116

Figure 5-5 Impact test set-up for EASP specimens 117

Figure 5-6 Impact test experimental set-up for RC specimens 119

Figure 5-7 High speed video footage of EASP1-C60 120

Figure 5-8 High speed video footage of EASP2-C60 121

Figure 5-9 High speed video footage of EASP1-C110 122

Figure 5-10 High speed video footage of EASP1-C110F 123

Figure 5-11 High speed video footage of EASP2-C110F 124

Figure 5-12 High speed video footage of RC-C110 125

Figure 5-13 High speed video footage of RC-C110F 126

Figure 5-14 EASP1 & EASP2 with C60 concrete after impact test 127

Figure 5-15 EASP1 & EASP2 with C110 concrete after impact test 127

Figure 5-16 EASP1 & EASP2 with C110F concrete after impact test 128

Figure 5-17 RC-C110 and RC-C110F specimens after impact test 128

Figure 5-18 EASP1H-C60 after impact test 129

Figure 5-19 EASP2H-C60 crater after the impact 130

Figure 5-20 EASP1H-C110 after impact test 130

Figure 5-21 EASP2H-C110 after impact test 131

Figure 5-22 EASP1H-C110F after impact test 131

Figure 5-23 Displacement time history of EASP2-C60 132

Figure 5-24 Displacement vs Time Graph of EASP1 specimens 134

Figure 5-25 Displacement vs Time graph of EASP2 specimens 134

Figure 5-26 Maximum displacement from image analysis (EASP1-C60) 135

Figure 5-27 Displacement vs Time graph of EASP1H specimens 136

Figure 5-28 Displacement vs Time graph of EASP2H specimens 136

Figure 5-29 EASP 1&2 (C60) strain measurement during 100 ms of impact 138

Figure 5-30 EASP 1&2 (C110) strain measurement during 100 ms of impact 138

Figure 5-31 EASP2-C110F strain measurement during 100 ms of impact 139

Figure 5-32 EASP 1H&2H (C60) strain measurement during 100 ms of impact 139

Figure 5-33 EASP 1H&2H (C110) strain measurement during 100 ms of impact 140

Figure 5-34 EASP 1C&2C (C110F) strain measurement during 100 ms of impact 141

Figure 5-35 Displacement-time history and cracking pattern of EASP1-C60 142

Figure 5-36 Displacement-time history and cracking pattern of EASP1-C110 142

Figure 5-37 Displacement-time history and cracking pattern of EASP1-C110F 143

Figure 5-38 Displacement-time history and cracking pattern of EASP2-C60 143

Figure 5-39 Displacement-time history and cracking pattern of EASP2-C110 144

Figure 5-40 Displacement-time history and cracking pattern of EASP2-C110F 144

Figure 5-41 Displacement-time history and cracking pattern of EASP1H-C60 145

Figure 5-42 Displacement-time history and cracking pattern of EASP1H-C110 145

Figure 5-43 Displacement-time history and cracking pattern of EASP1H-C110F 146

Figure 5-44 Displacement-time history and cracking pattern of EASP2H-C60 146

Figure 5-45 Displacement-time history and cracking pattern of EASP2H-C110 147

Figure 5-46 Displacement-time history and cracking pattern of EASP2H-C110F 147

Figure 5-47 FE simulation result of RC-C110 specimen 148

Figure 5-48 FE simulation result of RC-C110F specimen 149

Figure 5-49 RC-C110 vs EASP2-C110F after impact 152

Figure 6-1 Blast pressure time history 154

Figure 6-2 Stud to floor anchorage using steel angle 162

Figure 6-3 EASP1 panel 164

Figure 6-4 Blast test set-up of (A) EASP1 C110F and (b) EASP1 C110 165

Figure 6-5 High speed video footage of 5 kg TNT blast with vertical cylinder axis placement (jetting effect marked) 166

Figure 6-6 High speed video footage of 5 kg TNT blast with horizontal cylinder axis placement (jetting effect marked) 166

Figure 6-7 EASP1-C110F after blast test (perpendicular charge placement) 167

Figure 6-8 EASP1-C110 after blast test (parallel charge placement) 167

Figure 6-9 5kg TNT Charge (a) perpendicular placement and (b) parallel placement 168

Figure 6-10 Plan view of measurement lines for a cylindrical charge 169

Figure 6-11 Blast angle for (a) perpendicular charge (b) parallel charge 169

Figure 6-12 Equivalent spherical mass ratio at 22.5° and 0° for a cylindrical TNT 170

Figure 6-13 Blast loading curve for perpendicular charge 171

Figure 6-14 EASP1-C110F specimen blast simulation results 172

Figure 6-15 Equivalent mass ratio at 67.5°, 112.5° and 90° for cylindrical TNT 173

Figure 6-16 Blast loading curve for parallel charge 173

Figure 6-17 Displacement time history of EASP1-C110 specimen 175

Figure 6-18 Failure mode of the EASP1-C110F specimen 176

Figure 6-18 EASP1-C110 specimen blast test result 177

Figure 6-19 EASP1-C110 numerical model 177

2

sph

Subscripts:

ELFORM Element Formulation

SRA Shrinkage Reduction Admixture

Chapter 1 Introduction

Background of the study

In recent years, there have been many incidents of extreme events arising from bomb explosions ranging from Improvised Explosive Devices (IED) to Vehicle Borne Improvised Explosive Devices (VBIED) Table 1-1 list several bombing incidents that occurred between

1993 and 2012 which caused damage to buildings and resulted in a large number of human casualties

TAB LE 1-1 LIST OF B OMB IN G IN CID EN TS (W IKIP EDIA OR G, 2 012 )

Bombay, Maharashtra, India 13 car bombs (RDX) with shrapnel 257 killed, 713 injured Hotels, office, banks great damage to buildings Oklahoma, USA 3200 kg ammonium nitrate, 168 killed, 680 injured Alfred P Murrah Federal Building nitromethane, and tovex 324 buildings destroyed or damaged Khobar, Saudi Arabia mix of gasoline and explosive powder 20 killed, 372 injured Khobar Towers housing complex estimated to be equal to 9072 kg TNT destroyed structures Tanzania & Kenya Truck bomb 223 killed, more than 4000 injured United States embassies 900 kg of combined explosives nearby buildings collapsed Moscow, Russia 300-400 kg RDX 94 killed, 249 injured Apartment on 19 Guryanova Street section of the apartment collapsed New York City, USA Three large plane with lots of fuel Nearly 3000 killed World Trade Centre 3 buildings collapsed Bali, Indonesia Suicide backpack & large car bomb 202 killed, 240 injured Kuta district potassium chlorate, aluminum powder great damage to buildings Madrid, Spain IED (improvised explosive device) 191 killed, 2050 injured Madrid Commuter Train System

Mumbai, India RDX and ammonium nitrate 209 killed, 714 injured Mumbai Westen Line train combined with pressure cookers

Iraq Car bombs, fuel tanker 796 killed, 1562 injured Qahtaniya and Jazeera Almost 2 tons of explosives massive damage to buildings Shah Hasan Khel, Pakistan Suicide bombing 105 killed, 100+ injured Lakki Marwat district unknown explosives

Sana'a, Yemen Suicide bombing more than 120 killed, 350 injured Targeted to Yemeni Army parade unknown explosives

progressive collapse Typical modes of localized damage arise from the failure of one or a group of structural components due to the overpressure blast wave and flying debris generated during the explosion In some cases, progressive collapse follows when the failure of some critical structural elements occurs, eventually leading to the partial or total collapse of the entire structure This chain reaction of failure usually leads to extensive damage and heavy human casualties In addition, the flying fragments generated from primary sources (shrapnel packed together with the explosives) or secondary sources (fragmentation of materials after the blast wave) may contain enough energy to inflict further impact damage

One extreme bombing case involved the Alfred P Murrah Federal Building in Oklahoma City, USA which occurred on April 19, 1995 A rental truck packed with 3200 kg of combined ammonium nitrate, nitromethane, and tovex (equivalent to 2300 kg of TNT) detonated in front of the nine-story building The massive explosion blew off the building’s north wall, killing 168 people and injuring more than 680 people The blast wave also damaged more than 300 buildings in the immediate vicinity Figure 1-1 shows the collapsed part of the building after the explosion For safety reasons, the partially destroyed building was demolished after completion of site examination A new building sited north of the original site was built to replace the Murrah building designed with increased security measures (wikipedia.org, 2012)

FIGURE 1-1 CATASTROPHIC FAILURE OF ALFRED P MURRAH BUILDING

Briefly, the findings of the investigation committee concluded that the collapse began with the failure of only four columns located at the ground floor due to the blast wave produced

by the VBIED The absence of adequate loading transfer mechanisms from the upper floors

to the foundation causes part of the building to be sheared off before crashing to the ground (Osteraas, 2006)

Another bombing case in the city of Khobar, Saudi Arabia provides another example of the extensive damage produced from such blast events Estimated to be equal to 9072 kg of TNT, the mixture of gasoline and explosive powder packed into a truck was detonated adjacently to the eight storey structure housing US Air Force personnel It can be seen from Figure 1-2 that the eight storey building was half destroyed The damage seems to be similar to the Alfred P Murrah Federal building with the blast initiating failure of some of the first storey structural elements that led to more severe progressive collapse Such events occur very suddenly and usually results in a large number of casualties Hence, the failure of critical structural components, e.g first storey columns, must be avoided at all cost to mitigate the extent of localized failure and to guard against further progressive collapse

FIGURE 1-2 THE AFTERMATH OF KHOBAR HOUSING COMPLEX BOMBING

With increasing awareness especially in countries with a heightened threat level, engineers have been tasked to incorporate blast mitigation systems as protection for vulnerable buildings However, the main problem with such blast incidents is their unpredictability, both in terms of severity and timing The rare occurrence of such incidents also makes it difficult to quantify the relevant threat level Furthermore, the cost of incorporating adequate protective measures is usually prohibitive Due to these reasons, blast loadings are often not accounted for in the design of the vast majority of buildings

While highly protective structures are usually very expensive to build, a number of mitigation strategies may be adopted to upgrade protection level and lessen the undesirable consequences should such incidents occur Mitigation strategies such as enhancing structural robustness to provide adequate protection have been implemented both for upgrading the protection level of existing critical buildings and when new critical buildings are being built Moreover, a number of innovations and developments have been put forward, from using novel protective materials to the introduction of additional vehicle barriers targeting at mitigating the impact of bombing attacks

Mitigation by itself can be defined as an effort to reduce loss of property and life by lessening the consequences of disasters Different mitigation strategies will be needed for different types of structures For example, the strategy to be adopted in the case of new buildings should be considered early in the planning and design process This can include the adoption of suitable building layouts, robust security features and appropriate built-in protection systems In the case of existing buildings, the usual strategy will be to upgrade the protection level by adding physical barriers or installing additional protective sacrificial cladding through structural elements retrofitting Moreover, critical structures such as military and government buildings are usually accorded higher priority as they are usually prime targets requiring higher protection levels vis-à-vis ordinary civilian structures

The mitigation systems that can be implemented as blast protection can be grouped into three types The first system is the adoption of specially configured structural layouts, able

to channel the blast pressure wave away from critical structural elements The preconfigured layout could be through an open space or air tunnels installed to transfer part of the blast loading away from the critical structural elements The second system is

to add a protection layer to shield the critical structural component itself For example, the column may have an additional layer of protective material installed, e.g steel plate or fibre-reinforced polymer (FRP) to improve structural resistance The third system is to use protective panels robustly connected to the floor slabs, installed as a secondary or sacrificial protection layer (Figure 1-3) With this arrangement, the blast forces transmitted

to the frame from the panel will be resisted by the overall building inertia, rather than inducing local bending of the columns or other perimeter elements which might fail as a result (Hadden, 2003) With a gap provided between the structural components and the sacrificial panel, damage to the shielded structural vertical components (column or walls) can be minimized or even eliminated

FIGURE 1-3 TRANSMISSION OF BLAST LOAD FROM THE SECONDARY BARRIER

The third mitigation strategy in providing an additional layer of defence such as sacrificial cladding has several advantages The first is the flexibility in application The panel can be designed for various types of buildings to suit the parts of the structure they are to be attached, hence a drastic change in the building layout becomes usually unnecessary Second, the system can be customized to the different levels of protection required For example, the blast absorption rate may be increased by increasing the panel thickness or through changing the constituent materials used within the panel structure Other properties such as ductility or plastic deformation limits may be improved by using high ductility materials such as steel or FRP Third, the reparability of such sacrificial panels is relatively easier The damaged panel can be replaced quickly, especially if the connections are designed to allow for this function Last but not least, the cost of this system is expected

to be lower especially if such panels can be mass produced These advantages can be utilized to formulate a set of design criteria for the design of particular applications of such protective panel systems

Looking at the performance of protective technology nowadays, composite panels seems

to be a clear winner vis-a-vis ordinary reinforced concrete or solid steel plates on an equal weight basis The previously complicated fabrication processes involved in the production

of composite panels has been simplified with the advent of new fabrication technology resulting in stiffer panels without the need to introduce more materials For example,

Blast origin point Robust connection

panels with a honeycombed cellular structure as its core are able to provide the necessary structural rigidity and stiffness with a minimum amount of material This kind of panel is best suited for use as protective panels

It can be concluded that there is a need for mitigation systems that can deliver good performance at a reasonable cost for resisting blast loading and imbued with provisions for flexibility to suit a range of applications A secondary protection layer as sacrificial cladding has the potential to fulfil the requirements of such blast mitigation system The system satisfies the criteria of flexibility to be applied to most building types, customization ability to meet different protection levels desired and good repairability Composite panels seem to be the best choice in achieving the desired level of performance in protection The next section reviews some available composite panels for use against blast loading

Literature review

Composite panel as an energy absorber

The subsequent section review some studies conducted involving several types of composite panel structures that have been used as blast and impact protection panels or armour They include metal air sandwich panels or cellular metal panels, steel-concrete-steel sandwich panels, fibre composite panels, steel stud composite panels and multi-layer composite blast panels

Metal sandwich panels

Sandwich panels can be defined as a three-layer structure consisting of two thin outer skins

of high-strength material referred to as sandwich face plates, separated by a lightweight core material (Figure 1-4) They can be connected using a number of connection methods such as adhesive bonding, riveting, and welding The core material acts as a filler of the

core of the sandwich contributing to the overall panel strength and can be formed using various core configurations The combination of the core structure with outer skins creates

a panel with superior stiffness and lightweight characteristics The sandwich skin plates are designed to withstand bending and axial stresses whereas the cores are designed to resist transverse shear stress

FIGURE 1-4 SANDWICH PANEL TYPICAL GEOMETRY

The option to select a wide range of combinations of constituent materials for use as the core and skin plates as well as core structure configuration makes the sandwich panel fully customizable to meet the desired design requirements Until recently, the majority of sandwich panels have been designed with outer skins of fibre-reinforced polymer composites or thin sheet metals sandwiching a structural foam, timber or elastomeric cores The development of all-metal sandwich panels had been inhibited by the limitations associated with conventional cutting and welding processes that make fabrication of complex shapes and configurations difficult and expensive However, the new plasma and laser beam welding techniques have made such fabrication practical and cost effective (Haller et al., 2006)

FIGURE 1-5 METAL SANDWICH STRUCTURES (CEL-COMPONENTS, 2012)

The performance of all metal sandwich panels is often compared with that of solid metallic plates made of the same material, having equal weights Many core geometries are available for use as the core between the two outer skins of metal plates in a typical sandwich configuration Particular core geometry has specific stiffness and energy absorption characteristics Some can be very stiff, while others can be rather flexible, registering large deformations when the panels are subjected to blast or impulse loads

Guruprasad and Mukherjee (2000b) carried out experimental and numerical analyses of the behaviour of layered sacrificial claddings subjected to blast loading They found that the impulse transfer was reduced substantially at the distal face of the cladding because

of energy absorption through plastic deformation of the steel inside the core The analytical solutions seem to agree well with the experimental results The results suggest that properly designed sacrificial cladding may be very efficient in dissipating blasts

FIGURE 1-6 DEFORMED SHAPE FOR STEEL SANDWICH (GURUPRASAD & MUKHERJEE,

2000B)

Karr et al (2009) takes the research further by adding liquid encasement in the core structure similar to that used in the Guruprasad and Mukherjee (2000b) study Comparison between dry panels and wet panels were discussed, subjected to mechanical drop weight tests and blast tests Although the impulse transfers at the base remain essentially the same, the through-thickness collapse of the wet panel was substantially reduced The

addition of a liquid encased within the core increases the energy dissipation capacity of the sandwich panels

Xue and Hutchinson (2004) tested three core geometries of all metal sandwich panels: pyramidal truss, square honeycomb and folded panels The plates were clamped along their sides and subjected to a uniform impulsive load They found that all three types of sandwich plates were capable of sustaining a larger blast when compared with the corresponding solid plates of equal mass These seem to indicate that there is huge potential for the use of all metal sandwich panel construction for blast resistant structures McShane et al (2006) used metal foam projectiles (projectile made from deformable, aluminium foam structure, intended to simulate uniform shock loading, Figure 1-7) to test lattice cores sandwich panels Comparison of square-honeycomb and pyramidal core structure with monolithic plates of equal mass was discussed by quantifying the mid-span permanent transverse deflection It was found that sandwich panels with the square honeycomb structure outperform those with the pyramidal core

FIGURE 1-7 ALPORAS ALUMINIUM FOAM STRUCTURE (ALCARBON TECHNOLOGY

COMPANY)

In parallel with the Xue and Hutchinson (2004) in a metallic sandwich panels experimental study, Fleck and Deshpande (2004) developed an analytical method to calculate the dynamic response of metallic sandwich “beams” subjected to water and air blast Although the specimen’s core design is the same as panel tested by Xue and Hutchinson, the term

‘beam’ was used due to the smaller width to depth ratio The response of the sandwich beam was divided into three consecutive stages: the first stage involved the fluid-structure interaction problem, the second stage is the phase during which core compression controls, and in the third stage, the clamped beam specimen is brought to rest by plastic stretching and bending They concluded that for both air and water blast, the diamond celled core sandwich beam gave the best performance due to the higher stiffness provided

by the core This study clearly shows excellent results based on the performance of steel sandwich panels tested compared with a corresponding solid metal plate of equal weight

FIGURE 1-8 MAXIMUM BLAST IMPULSE SUSTAINED BY MONOLITHIC BEAMS (FLECK &

DESHPANDE, 2004)

Figure 1-8 shows a numerical comparison of steel sandwich clamped beams as reported

by Fleck and Deshpande (2004) The graph compares the non-dimensional total mass of the steel sandwich structures and the maximum impulse that can be absorbed It can be

seen that the diamond celled core sandwich panels have the highest energy absorption capacity followed by those with the square honeycomb core Other core configurations also show good potential for use as energy absorbers Zhu et al (2009) also carried out an experimental investigation on honeycomb core and aluminium foam core metallic sandwich panels subjected to blast loading The same three phases of deformation was observed, as reported by the Fleck and Deshpande study However, no comparison was made with monolithic metallic plates of equal mass

In general, a well-designed sandwich panel can sustain a significantly larger blast impulse when compared with the corresponding solid metal plate of equal weight This can help to mitigate the level of damage experienced by the component when exposed to blast loadings High stiffness properties with relatively much lower mass have acknowledged benefits in critically lightweight structures such as those used in the aerospace industries

Steel-concrete-steel (SCS) sandwich panels

The first steel-concrete-steel (SCS) sandwich construction was proposed for the Conwy submerged tube tunnel construction (Tomlinson et al., 1989) In the first generation of SCS panels, the outer plates were connected by welding overlapping headed shear studs This double-skin sandwich construction (DSC) proved difficult to implement due to the positioning pattern of the shear connectors adopted and the lack of control of concrete thicknesses during fabrication The advantages of this system are that the external steel plates act as primary reinforcement and permanent formwork as well as providing a high resistant barrier to absorb blast and impact

In SCS panel construction, mechanical shear connectors are essential in developing composite action between the external steel plates and the concrete core The first panel developed utilized overlapping welded studs This design seems to be susceptible to

cracking and premature vertical shear failure in some cases Hence, a set of guidance and recommendations regarding the maximum load permitted has been postulated to prevent such failure modes (Oduyemi & Wright, 1989) Bi-Steel as an alternative construction method overcame some of the construction problems of the DSC panels The prefabrication technique developed by British Steel (Bowerman et al., 2002) allows both ends of a shear connector to be simultaneously welded by friction to the steel face plates Since the fabrication required special techniques that may not be readily available, the usage of such systems is limited Another SCS sandwich construction with innovative J-hook connectors as shear connectors has also been proposed recently as an alternative low-cost construction technique (Liew & Wang, 2007) The J-hook connectors require very basic construction tools, readily available in practice Extensive research had also been done to understand the behaviour of the J-hook panels when subjected to static and impact loading They showed considerably better performance compared with other non-composite systems (Dai & Liew, 2010; Liew et al., 2009)

FIGURE 1-9 BI-STEEL SCS PANEL (TATA-STEEL, 2012)

The results from previous experimental studies clearly show the superiority in the performance of steel-concrete sandwich panels SCS with various shear connectors and design construction has been proposed, researched, and developed and the trend now is towards newer innovative types of shear connectors and concrete However, some design issues need to be addressed One of them is the load slip relationship at the steel plate - concrete interface Many have concluded that the strength of a typical SCS panel is

dependent on the shear strength at the steel-concrete interface and this interfacial shear strength is primarily influenced by the concrete properties as used in the core (Dai & Liew, 2010; Ollgaard et al., 2003) The low tensile strength of concrete has all along been a persistent design issue The utilization of fibre reinforced concrete may help to mitigate this problem as fibre incorporation greatly increases the tensile capacity of the concrete

Although it has been known that steel concrete sandwiches can perform better compared with each of its constituent materials on their own (e.g steel or concrete only) of similar weight, the utilization of concrete or steel as a standalone material for structural protection against blast loads is still popular This may be due to the limitations associated with existing analytical and numerical modelling of steel concrete composites, especially if design codes and suitable guidelines are lacking However, it is expected that as technology progresses, such limitations may be overcome, and composites with improved properties

of the respective constituents can be tapped to achieve enhanced performance against blast and impact loading

Fibre composite panels

Fibre reinforced concrete has been successfully utilized in many applications such as precast products, architectural panels, high-security buildings and hydraulic structures It was invented to address to inherent weaknesses associated with the poor tensile strength

of plain concrete Plain concrete cracks easily, allowing easy access to detrimental substances to decrease service life The principle behind fibre reinforced concrete is to reinforce plain concrete that is brittle with randomly distributed short, discontinuous fibres made of various materials Mixtures of one concrete with different fibre types and fibre percentage will result in different static and dynamic performance of the fibre reinforced concrete produced This has resulted in a number of novel and innovative applications, especially in enhancing blast and impact loading resistance

In ordinary reinforced concrete, the moment capacity will drop abruptly when initial cracking occurs The weakest link usually occurs at the interfacial transition zone (ITZ) or the interface between aggregate and cement matrix (Zampini et al., 1995) By introducing fibres, the cracks may be bridged, delaying further crack propagation, leading to higher post crack ductility and greater residual capacity

FIGURE 1-10 BRIDGED CRACK IN FRHSC (BRIGHTHUB-ENG., 2011)

Incorporation of various types of fibres improves the mechanical properties of concrete, increase the resistance to cracking and crack propagation, provides required ductility, and increases the energy absorption characteristics of concrete It was found that concrete with polymer fibres exhibited an increase in energy absorption capacity or toughness at high strain rate loading (Bindiganavile & Banthia, 2001) These properties are important and expected to have a significant effect on the response of the fibre reinforced concrete when subjected to blast loading

Magnusson (2006) tested steel fibre reinforced concrete beams subjected to static and dynamic loads The dynamic load was generated by explosive charge detonation The concrete compressive strength varied between 36 to 189 MPa with 1% steel fibre content

by volume The experimental results show a significant increase in the load capacity when the test specimens were subjected to air blast loading testing, with observed strength increasing by around 30-170% with respect to the static test However, the toughness was

reduced at higher concrete strength This could be explained by an increase in concrete brittleness in the case of higher strength concrete Moreover, in the study on the effects

of strain rate of impact loading using a drop-weight impact machine by Naaman and Gopalaratnam (1983), it was found that the energy absorbed by the fibre reinforced composite under static loading can be up to two orders of magnitude higher than that of

a similar unreinforced matrix The impact test results showed even more significant increases, up to three-fold, in the modulus of rupture and energy absorption by the composite when the strain rate increased from 0.5e-5 to 1.2 strains per second The performance of fibre reinforced composite concrete panels is proven to be greater than plain concrete with equal characteristic strength; hence it is suitable for use in blast and impact mitigation panels

Extensive tests conducted on various composite structural panels including conventional reinforced concrete panels, steel fibre reinforced concrete panels, profiled steel sheeting reinforced concrete panels, steel-air-steel sandwich panels and steel-concrete-steel sandwich panels subjected to explosive loading were conducted by Lan et al (2005) Seventy-four specimens were tested with charge weights ranging between 8 to 100 kg with

a 5 m standoff distance Comparing the reinforced concrete panel with and without steel fibres, showed a positive effect associated with fibre incorporation Reduced panel deflection and material fragmentation were observed Moreover, the fibre reinforced concrete panel with 1% fibre by weight could overcome panel breaching

Wu, et.al.(2009) conducted further research of ultra-high performance concrete materials subjected to cylindrical shaped charge explosions A series of test were done with both unreinforced and reinforced ultra-high performance fibre concrete (UHPFC) panels reinforced with externally bonded fibre reinforced polymer (FRP) plates and normal reinforced concrete panels (NRC) The blast test used various charge weights and standoff