Advanced Materials and Structures

and their Fabrication Processes

Book manuscript, Narvik University College, HiN

Dag Lukkassen and Annette Meidell

August 23, 2007

Contents

I Materials and Structures 8

1 Ceramics and glasses . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.1 Ceramics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.2 Glasses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.3 Industrial Importance of Ceramics and Glasses . . . . . . . . . . . 13
1.4 Industrially Important Glasses . . . . . . . . . . . . . . . . . . . . 14
1.5 Technical Ceramics . . . . . . . . . . . . . . . . . . . . . . . . . . 14
1.5.1 High-performance ceramics . . . . . . . . . . . . . . . . . . 15
2 Metals and metal alloys . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.1 High-peformance Metals . . . . . . . . . . . . . . . . . . . . . . . 19
2.1.1 High-Performance Aluminum . . . . . . . . . . . . . . . . 19
2.1.2 High-Performance Steel . . . . . . . . . . . . . . . . . . . . 20
2.1.3 Space-age metals . . . . . . . . . . . . . . . . . . . . . . . 20
3 Polymers and Plastics . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
3.1 The structure of polymers . . . . . . . . . . . . . . . . . . . . . . 23
3.2 Crystallinity in polymers . . . . . . . . . . . . . . . . . . . . . . . 25
3.2.1 The glass transition temperature and melting temperature 26
3.3 Plastics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.4 Properties of plastics . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.4.1 Thermosettings . . . . . . . . . . . . . . . . . . . . . . . . 30
 3.4.2 Types of Thermosettings . . . . . . . . . . . . . . . . . . . 30
3.4.3 Thermoplastics . . . . . . . . . . . . . . . . . . . . . . . . 32
3.4.4 Types of Thermoplastics . . . . . . . . . . . . . . . . . . . 34
3.5 Classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
3.6 Additives in Plastics . . . . . . . . . . . . . . . . . . . . . . . . . 39
3.6.1 Oriented Plastics . . . . . . . . . . . . . . . . . . . . . . . 41
3.7 Elastomers & Rubbers . . . . . . . . . . . . . . . . . . . . . . . . 41
3.7.1 Rubber and Artificial Elastomers . . . . . . . . . . . . . . 43
3.8 Biopolymers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
3.8.1 What makes a polymer a biopolymer? . . . . . . . . . . . 47
3.8.2 Applications of biopolymers . . . . . . . . . . . . . . . . . 48
3.8.3 Properties of biopolymers . . . . . . . . . . . . . . . . . . 48
3.8.4 Additional information on raw materials in biopolymers . . 50
3.8.5 Plastic taste better with sugar . . . . . . . . . . . . . . . . 51
4 Composites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
4.1 The history of composites . . . . . . . . . . . . . . . . . . . . . . 56
4.2 Natural composites . . . . . . . . . . . . . . . . . . . . . . . . . . 56
4.3 Man-made Composites . . . . . . . . . . . . . . . . . . . . . . . . 58
4.3.1 Fiber-Reinforced Composites . . . . . . . . . . . . . . . . . 60
4.3.2 Classification of composites . . . . . . . . . . . . . . . . . 60
4.4 Ceramic matrix composites CMC . . . . . . . . . . . . . . . . . . 61
4.5 Metal Matrix Composites MMC . . . . . . . . . . . . . . . . . . . 64
4.6 Polymer Matrix Composites PMC . . . . . . . . . . . . . . . . . . 65
4.7 Fiber reinforcement . . . . . . . . . . . . . . . . . . . . . . . . . . 66
4.7.1 Choosing fibers . . . . . . . . . . . . . . . . . . . . . . . . 67
4.7.2 Glass fibers . . . . . . . . . . . . . . . . . . . . . . . . . . 69
4.7.3 Carbon fibers . . . . . . . . . . . . . . . . . . . . . . . . . 70
4.7.4 Boron fiber . . . . . . . . . . . . . . . . . . . . . . . . . . 71
4.7.5 Ceramic fibers . . . . . . . . . . . . . . . . . . . . . . . . . 71
4.7.6 Polymer Fibers . . . . . . . . . . . . . . . . . . . . . . . . 72
4.7.7 Carbon nanotube (CNT) fibers . . . . . . . . . . . . . . . 74
4.7.8 Fiber hybrids . . . . . . . . . . . . . . . . . . . . . . . . . 75
4.7.9 Spider silk . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
4.8 Aerogel composites . . . . . . . . . . . . . . . . . . . . . . . . . . 76
4.9 Textile composites . . . . . . . . . . . . . . . . . . . . . . . . . . 78
4.10 Bio-inspired materials . . . . . . . . . . . . . . . . . . . . . . . . 78
4.11 Self-healing composites . . . . . . . . . . . . . . . . . . . . . . . . 79

2
 4.12 Left-handed metamaterials . . . . . . . . . . . . . . . . . . . . . . 81
5 Smart (intelligent) Materials and Structures . . . . . . . . . . . . . . . 87
5.1 Color Changing Materials . . . . . . . . . . . . . . . . . . . . . . 88
5.1.1 Photochromic materials . . . . . . . . . . . . . . . . . . . 88
5.1.2 Thermochromic materials . . . . . . . . . . . . . . . . . . 88
5.2 Light Emitting Materials . . . . . . . . . . . . . . . . . . . . . . . 88
5.2.1 Electroluminescent materials . . . . . . . . . . . . . . . . . 89
5.2.2 Fluorescent materials . . . . . . . . . . . . . . . . . . . . . 89
5.2.3 Phosphorescent materials . . . . . . . . . . . . . . . . . . . 90
5.3 Moving Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
5.3.1 Conducting polymers . . . . . . . . . . . . . . . . . . . . . 91
5.3.2 Dielectric elastomers . . . . . . . . . . . . . . . . . . . . . 92
5.3.3 Piezoelectric materials . . . . . . . . . . . . . . . . . . . . 92
5.3.4 Polymer gels . . . . . . . . . . . . . . . . . . . . . . . . . . 93
5.3.5 Shape memory materials (SMM) . . . . . . . . . . . . . . 94
5.3.6 Nanostructured Shape Memory Materials . . . . . . . . . . 95
5.3.7 Magnetic Shape Memory (MSM) Materials . . . . . . . . . 96
5.4 Temperature Changing Materials . . . . . . . . . . . . . . . . . . 97
5.4.1 Thermoelectric materials . . . . . . . . . . . . . . . . . . . 97
6 Functional Gradient Materials . . . . . . . . . . . . . . . . . . . . . . . 98
7 Solar Cell Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
7.1 Light-absorbing materials . . . . . . . . . . . . . . . . . . . . . . 101
7.2 Silicon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
7.3 Thin films . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
7.4 Silicon solar cell device manufacture . . . . . . . . . . . . . . . . . 104
8 Nano - materials - and technology . . . . . . . . . . . . . . . . . . . . . 106
8.1 Carbon Nanotubes (CNT) . . . . . . . . . . . . . . . . . . . . . . 107
8.1.1 Strength . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
8.2 Nanocomposites . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
8.3 Flexible ceramics . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
8.4 New type of High-performance Ceramic . . . . . . . . . . . . . . . 111
8.5 Back to square one? . . . . . . . . . . . . . . . . . . . . . . . . . 112
9 Cellular Solids, Structures & Foams . . . . . . . . . . . . . . . . . . . . 116
9.1 Metal Foams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
9.1.1 Aluminium foam: . . . . . . . . . . . . . . . . . . . . . . . 118
9.2 Polymeric foam . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
9.3 Refractory foams / Ceramic foam . . . . . . . . . . . . . . . . . . 122

3
 9.3.1 Carbon foam . . . . . . . . . . . . . . . . . . . . . . . . . 123
9.4 Hierarchical structures (multiscale structures) . . . . . . . . . . . 123
9.5 Biomaterials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124

II Sandwich Constructions 127

10 Why use sandwich constructions? . . . . . . . . . . . . . . . . . . . . . 128
10.1 Face materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
10.2 Core materials and structures . . . . . . . . . . . . . . . . . . . . 131
10.2.1 Foam Cores . . . . . . . . . . . . . . . . . . . . . . . . . . 132
10.2.2 Honeycomb Cores . . . . . . . . . . . . . . . . . . . . . . . 132
10.2.3 Corrugated Cores . . . . . . . . . . . . . . . . . . . . . . . 135
10.2.4 Wood Cores . . . . . . . . . . . . . . . . . . . . . . . . . . 136
10.3 Adhesives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
11 Design of sandwich constructions . . . . . . . . . . . . . . . . . . . . . 140
11.1 Design of sandwich beams . . . . . . . . . . . . . . . . . . . . . . 140
11.2 Preliminaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
11.3 The Flexural Rigidity and Shear Rigidity . . . . . . . . . . . . . . 143
11.4 Tensile and Compressive Stresses . . . . . . . . . . . . . . . . . . 144
11.4.1 Due to bending (transversal loading) . . . . . . . . . . . . 144
11.4.2 Due to in-plane loading (axial) . . . . . . . . . . . . . . . 147
11.5 Shear stresses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
11.5.1 Summary of approximations . . . . . . . . . . . . . . . . . 153
11.6 Sandwich design: stiffness, strength and weight . . . . . . . . . . 155
11.7 Example of beam calculations . . . . . . . . . . . . . . . . . . . . 157
11.8 Strength and stiffness design example . . . . . . . . . . . . . . . . 160
12 Failure modes of sandwich panel . . . . . . . . . . . . . . . . . . . . . . 163
12.1 Failure loads and stresses . . . . . . . . . . . . . . . . . . . . . . . 164
12.2 Failure-mode maps . . . . . . . . . . . . . . . . . . . . . . . . . . 167
12.2.1 Transition equation between face yielding and face wrinkling 167
12.2.2 Transition equation between face yield and core shear . . . 169
12.2.3 Transition equation between face wrinkling and core shear 169
13 Design Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174
13.1 The stiffness of sandwich structures and its optimization . . . . . 174
13.1.1 Example of minimum weight design for given stiffness . . . 177

4
III Cellular Solids 180
14 Some definitions of cellular solids . . . . . . . . . . . . . . . . . . . . . 180
14.1 Mechanics of honeycombs . . . . . . . . . . . . . . . . . . . . . . 183
14.2 In-plane deformation properties, uniaxial loading of hexagonal hon-
eycombs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
14.2.1 Linear-elastic deformation . . . . . . . . . . . . . . . . . . 184
14.3 Out-of-plane deformation properties . . . . . . . . . . . . . . . . . 187
14.3.1 Linear-elastic deformation . . . . . . . . . . . . . . . . . . 188

IV Mechanics and Effective Properties of Composite

Structures and Honeycombs 191
15 On effective Properties of Composite Structures . . . . . . . . . . . . . 191
16 The thermal problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
17 Isotropic elastic materials . . . . . . . . . . . . . . . . . . . . . . . . . . 193
18 Orthotropic composites . . . . . . . . . . . . . . . . . . . . . . . . . . . 194
19 Square symmetric unidirectional two-phase structure . . . . . . . . . . 196
19.1 Calculation of stiffness and compliance matrix . . . . . . . . . . . 198
20 Numerical methods for periodic structures . . . . . . . . . . . . . . . . 200
20.1 Coordinate transformation . . . . . . . . . . . . . . . . . . . . . . 202
21 A computational example . . . . . . . . . . . . . . . . . . . . . . . . . 206

V Fabrication processes 210

22 Fabrication of plastics . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
22.1 Processing of Rubber and elastomers . . . . . . . . . . . . . . . . 218
23 Fabrication of Fiber-Reinforced Composites (FRC) . . . . . . . . . . . 221
23.1 Manufacturing processes of Reinforcements . . . . . . . . . . . . 221
23.2 Prepregs, Preforms and Compounds . . . . . . . . . . . . . . . . . 223
23.3 Fabrication processes of FRC . . . . . . . . . . . . . . . . . . . . 224
23.4 Fabrication of Functionally Gradient Materials . . . . . . . . . . . 229
23.5 Fabrication of sandwich constructions . . . . . . . . . . . . . . . . 229
24 Layer Manufacturing Technology (LMT) . . . . . . . . . . . . . . . . . 233
24.1 Ballistic particle manufacturing (inkjet) BMP . . . . . . . . . . . 235
24.2 Fused deposition modelling - FDM . . . . . . . . . . . . . . . . . 235
24.3 LOM - Laminated object manufacturing . . . . . . . . . . . . . . 236
24.4 Selective laser sintering - SLS . . . . . . . . . . . . . . . . . . . . 236

5
 24.5 Laser Engineered Net Shaping (LENS) . . . . . . . . . . . . . . . 237
24.6 Stereo lithography - SLA . . . . . . . . . . . . . . . . . . . . . . . 238
24.7 Solid ground curing - SGC . . . . . . . . . . . . . . . . . . . . . . 239
24.8 Three dimensional printing (3DP) . . . . . . . . . . . . . . . . . . 240
24.9 PowderProcessing/PowderMetallurgy . . . . . . . . . . . . . . . . 241
24.10Vapor Deposition CVD/PVD . . . . . . . . . . . . . . . . . . . . 243
24.11Selection of a layered manufacturing process . . . . . . . . . . . . 243
25 Design Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . 247
25.1 Designing for Manufacturability (DFM) ) . . . . . . . . . . . . . . 247
25.2 Product design guidelines . . . . . . . . . . . . . . . . . . . . . . 248
25.3 Evaluation of design alternatives . . . . . . . . . . . . . . . . . . . 249
25.4 The meaning of colors . . . . . . . . . . . . . . . . . . . . . . . . 250

6
Preface
This manuscript is written in order to use it as lecture notes in the course ”Ad-
vanced materials” given at Narvik University College for master students in the
field of Engineering Design. In the course the students will be familiar with differ-
ent kind of advanced materials and structures such as for instance smart materials,
functional gradient materials, polymers and plastics, nano-materials, elastomers
and rubbers, biopolymers, cellular solids and structures, and different types of
composite materials, which can be found in Part 1 of this report. In Part 2, the
fabrication-processes of different advanced materials, structures and forms is de-
scribed. The different types of sandwich constructions included the design of such
structures are subjects that are discussed in Part 3, while in Part 4 we will have
a closer look at mechanics and effective properties of composite structures and
honeycombs.
This report may also be used as literature for students studying mechanical
engineering, material science, production engineering. We assume that the reader
has a bachelor degree in Engineering.

7
 Figure 0.1: Classes of materials.

Part I
Materials and Structures
In this part we will mainly consider the concepts of materials and structures. The
difference between a material and a structure is not clearly defined. Many draw
the lines between what you understand as a homogeneous material when you see
it with your bare eyes, and the inhomogeneous material structure that you clearly
see is made up of a fixed geometry or mixing of materials. For instance an alloy is
by this definition a material even though it consists of two or more components,
but a honeycomb core built up of two different components is a structure.
Materials are often classified into the six broad classes that are shown in figure
0.1; metals, ceramics, glasses, elastomers, polymers and composites. But, when
we also include material structures, the number is bigger, and the classification
of the term ”materials and structures”, even though it is not a conventional way
of making a classification, may look like the one purposed in figure 0.2. We will
in this book briefly mention main properties of each group in figure 0.2, but the
advanced term in the title of the book refers to a thoroughly study of composites,
polymers, smart materials, sandwich constructions, nano-technology, functional
materials, cellular structures.

8
 Figure 0.2: Classes of materials and structures.

High Performance Materials (HPM)

High-performance materials (or Advanced Engineering Materials) are materi-
als that provide specific performance advantages in comparison with the counter-
part conventional materials. Often it is difficult to place materials strictly into the
group of high-performance group or other groups, but we often divide materials
into the following main groups:
• Standard materials, which are used in products that is exposed to non-
critical environments and low-stress applications
• Standard Engineering Materials, which are used in products that must
have general bearing and wear properties
• High-performance materials or advanced engineering materials, which
are used in products that must have superior properties (extreme service environ-
ments, superior chemical resistance, wear resistance, and loading properties)

There are several types of materials currently called high-performance mate-

rials, for instance high-performance concrete, high-performance composites, high-
performance plastics, high-performance aluminum, high-performance ceramics,
and high-performance steel. What they all have in common is that they have
outstanding properties compared to the materials we used earlier. In short time

9
maybe the materials we know as HPM today may not be so high-performance
tomorrow, since the materials-science is rapidly changing and growing.
While research laboratories are still exploring ways to exploit these materi-
als, some of them are ready for use. Because there is little data available on the
long-term results of many high-performance materials and because there is fre-
quently a relatively high initial cost, states are reluctant to take on such a venture
independently.

Questions:
What is the conventional way of making a classification of materials?
What is the new way of making a classification of materials?
What is so special about high performance materials?

References:
High-performance materials:
http://www.hiper-group.com/hcproducts.htm
http://www.greatachievements.org/?id=3809

free internet books:

http://books.nap.edu/

10
1. Ceramics and glasses
A ceramic is often broadly defined as any inorganic nonmetallic material. By this
definition, ceramic materials would also include glasses; however, many materials
scientists add the stipulation that ”ceramics” must also be crystalline. Recall
that crystalline materials have their molecules arranged in repeating patterns, see
Figure 1.1. Therefore we often divide the ceramics and glasses into two separate

Figure 1.1: Crystalline structure.

subgroups of materials.

1.1. Ceramics
Examples of ceramic materials can be anything from NaCl (table salt) to clay
(a complex silicate).The term ceramic comes from the Greek word keramikos,
which means burnt stuff, indicating that desirable properties of these materials
are normally achieved through a high-temperature heat treatment process called
firing. Ceramics are refractory (fireproof) materials, and has therefore a melting
◦
temperature > 1580 C.
Due to the covalent character of the chemical bond refractory ceramics exhibit
very high (about 3000◦ C) melting point, low mobility of atoms, low plasticity
and high hardness at temperatures up to 2000◦ C. Therefore, refractory materi-
als can substitute metals, alloys and intermetallics in a lot of high temperature
engineering, chemical and electronic applications.
Ceramics are in general hard, brittle, high-melting-point materials with low
electrical and thermal conductivity, low thermal expansion, good chemical and
thermal stability, good creep resistance, high elastic modulus, high compressive
strength, low density, high stiffness, high hardness, high wear resistance, and high

11
corrosion resistance. Many ceramics are good electrical and thermal insulators.
Some ceramics have special properties: some ceramics are magnetic materials;
some are piezoelectric materials; and a few special ceramics are superconductors
at very low temperatures. Ceramics are widely used in the electrical industry
mostly due to their high electrical resistance. Ceramics and glasses have one
major drawback: they are brittle. Their extremely low fracture toughness is the
drawback of ceramics in comparison with metals. It means that ceramics have a
very low tolerance of crack-like flaws.
Ceramic fibers such as graphite and aluminum oxide with their extremely high
stiffness have led to the production of fiber-reinforced composites. These materials
are only a few of an ever-growing list of industrially important ceramics. Recently,
new groups of ceramics have emerged with the possibility to use them as load-
bearing materials.

1.2. Glasses
Glass is an inorganic nonmetallic material that does not have a crystalline struc-
ture. Such materials are said to be amorphous. Recall that amorphous materials
have their molecules arranged randomly and in long chains which twist and curve

Figure 1.2: Amorph structure.

around one-another, making large regions of highly structured morphology un-

likely, see Figure 1.2. Examples of glasses range from the soda-lime silicate glass
in soda bottles to the extremely high purity silica glass in optical fibers.
The major raw material of glass is sand (or "quartz sand") that contains
almost 100 % crystalline silica in the form of quartz. Most glass formulations
contain about 70—72 % by weight of silicon dioxide (SiO2). Soda-lime glass which
is the most common form of glass contains nearly 30 % sodium and calcium oxides

12
 Figure 1.3: A transmission cable containing hundreds of glass optical fibers.

or carbonates. Pyrex is borosilicate glass containing about 10 % boric oxide. Lead

crystal is a form of lead glass that contains a minimum of 24 % lead oxide.
Large natural single crystals of quartz are pure silicon dioxide, and upon crush-
ing are used for high quality specialty glasses. Synthetic amorphous silica, an
almost 100 % pure form of quartz, is the raw material for the most expensive
specialty glasses.

1.3. Industrial Importance of Ceramics and Glasses

Silicon [Si] (do not confuse it with Silicone which are mixed inorganic-organic
polymers) has many industrial uses, for instance is it the principal component
of most semiconductor devices (whose electrical conductivity is in between that
of a conductor and that of an insulator), most importantly integrated circuits or
microchips. Silicon has been THE material which has made computers possible.
In the form of silica and silicates, silicon forms useful glasses, cements, and
ceramics. It is also a component of silicones, a class-name for various synthetic
plastic substances made of silicon, oxygen, carbon and hydrogen, often confused
with silicon itself.
Glasses have historically been used for low technology applications such as
soda bottles and window panes. However, glasses, like ceramics, have recently
found new application in high technology fields (particularly the semiconductor
microelectronics industry where silica is widely used as an insulator in transistors
and the fiber optic cable industry where high purity silica glass has made advanced
telecommunications possible, see Figure 1.3).
As with ceramics, the list of industrially important glasses also continues to
grow. As a result of their unique properties, ceramics and glasses are materials
which will be used extensively in areas such as aerospace, automobiles, microelec-
tronics, and telecommunications.

13
1.4. Industrially Important Glasses
Silicon [Si] (do not confuse with Silicone which are mixed inorganic-organic poly-
mers) has many industrial uses, for instance is it the principal component of
most semiconductor devices (whose electrical conductivity is in between that of
a conductor and that of an insulator), most importantly integrated circuits or
microchips.
In the form of silica and silicates, silicon forms useful glasses, cements, and
ceramics. It is also a component of silicones, a class-name for various synthetic
plastic substances made of silicon, oxygen, carbon and hydrogen, often confused
with silicon itself.

• Silica glass (SiO2 ) is used for optical fibers when it is very pure.

• Soda-lime glass (SiO2 -Na2 O-CaO) is the standard glass used for bottles and
windows due to its low cost and easy manufacturing.

• Borosilicate glass (SiO2 -B2 O3 ) is good in applications where thermal shock

resistance is necessary (e.g. laboratory glassware) because of its low coeffi-
cient of thermal expansion

• Lead glass (SiO2 -PbO) commonly known as ”crystal”, this glass has a high
index of refraction causing it to sparkle (much like a diamond)

1.5. Technical Ceramics

Technical Ceramics covers ceramic materials and products for technical applica-
tion. Terms such as:
· functional ceramics (Components which fulfill an electrical, magnetic, dielectri-
cal, optical etc. function)
· structural ceramics, engineering ceramics, or industrial ceramics (Components
which are subjected mainly to mechanical loads.)
· electro-ceramics
· cutting ceramics
· bio-ceramics
· high-performance ceramics (see below)
describe the products groups of Technical Ceramics, but due to overlapping
there is not a clear classification.

14
 Figure 1.4: Products made of ceramic material.

1.5.1. High-performance ceramics

High-performance ceramics are ceramics with incredibly light weight, high hard-
ness, non-corrodable, high melting points, high price and advanced applications,
see Figure 1.4.
Components made from high-performance ceramics are often determining the
functionality of machinery by possessing excellent corrosion resistance, very high
thermal stability and wear resistance, as well as high biocompatibility. Density,
porosity, electrical and optical characteristics can be varied within a wide range.
Highly specialized and flexible production techniques enable us to match our prod-
ucts to the specific customer demands.
In ENV 12212 (classification system for European comity for standardiza-
tion, CEN) high-performance ceramics are defined as ”highly developed, high-
performance applicable ceramic material, which is mainly non-metallic and inor-
ganic, and has certain functional properties.” The term is seen as a differentiation
to traditional clay based ceramic that includes china-ware, sanitary ceramic tiles
and bricks and covers all technical ceramics. The materials for electrical engineer-
ing are standardized according to the international standard IEC 627 (Interna-
tional Electrotechnical Commission, IEC).

The terms used above are often still used in classify technical ceramics. How-
ever a precise classification is only possible if the materials are listed under their
following chemical composition:

• silicate ceramics

— technical porcelain (quartz, feldspar and kaolin. )

15
 — steatite (major component: soapstone, additives: clay and flux)
— cordierite (these magnesium silicates occur during the sintering of soap-
stone with added clay, kaolin, corundium and mullite.)
— mullite-ceramic (Al2 O3 and SiO2 ,where mullite: (3 Al2 O3 2 SiO2 ) and
corundium (Al2 O3 ) )

• oxide ceramics

— aluminium oxide (AL2 O3 )

— magnesium oxide (MgO)
— zirconium oxide (ZrO2 )
— aluminium titanate (AT)
— piezo ceramic (PZT), (the most important piezoelectrical ceramic ma-
terials are based on the oxide mixed crystals system lead zirconate and
lead titanate)

• nonoxide ceramics:

— carbide: Silicon carbide (SIC), Sintered silicon carbide (SSIC), Reac-

tion bonded silicom infiltrated silicon carbide (SSIC), Recrystallized
silicon carbide (RSIC), and Nitride bonded silicom carbide (NSIC)
— nitride: Silicon nitride (Si3 N4 ), Silicon aluminium oxynitride (SIALON),
and Aluminium nitride (ALN).

Questions:
• What is the special properties of ceramics and glasses?
• What are they used for?
• What are the industrally important glasses?
• What kind of different ceramics are there?
• What is a high-performance ceramic?
• What is the drawback of ceramics compared to metals?
References:
http://www.keramverband.de/keramik/englisch/fachinfo/
werkstoffe/definitionen.htm

16
http://www.globaltechnoscan.com/
31stOct-6thNov02/high_performance_ceramics.htm
http://www.globaltechnoscan.com/
http://www.sciam.com/explore_directory.cfm
http://www.mse.cornell.edu/courses/engri111/ceramic.htm
http://www.mse.cornell.edu/courses/engri111/impglass.htm
http://biotsavart.tripod.com/327.htm
http://www.engr.sjsu.edu/WofMatE/

17
 Figure 2.1: A suspension bridge.

2. Metals and metal alloys

Metals are elements that generally have good electrical and thermal conductivity.
Many metals have high strength, high stiffness, and have good ductility. Some
metals, such as iron, cobalt and nickel are magnetic. Many metals and alloys have
high densities and are used in applications which require a high mass-to-volume
ratio.
At extremely low temperatures, some metals and intermetallic compounds be-
come superconductors (a superconductor can conduct electricity without electrical
resistance at temperatures above absolute zero. The change from normal electrical
conductivity to superconductivity occurs suddenly at a critical temperature Tc ).

Pure metals are elements which comes from a particular area of the periodic
table. Examples of pure metals include copper in electrical wires and aluminum
in cooking foil and beverage cans.
Metal Alloys contain more than one metallic element. Their properties can be
changed by changing the elements present in the alloy. Examples of metal alloys
include stainless steel which is an alloy of iron, nickel, and chromium; and gold
jewelry which usually contains an alloy of gold and nickel.
Some metal alloys, such as those based on aluminum, have low densities and
are used in aerospace applications for fuel economy. Other examples: Many metal
alloys also have high fracture toughness, which means they can withstand im-
pact and are durable. Many beverage cans are made of aluminum metal. Many
structures, such as this suspension bridge, are made of steel alloys, see Figure
2.1. Aircraft skins are made of lightweight aluminum alloys with high fracture
toughness.

18
2.1. High-peformance Metals
When metal components must perform under critical conditions, manufacturers
need high-performance metals. Titanium-base alloys, and specialty steels for the
aerospace industry are metals of exceptional wear resistance, corrosion resistance,
heat resistance, toughness, and strength. Nickel- and cobalt-based materials are
known as superalloys. Turbine blades are for instance made of superalloys in
◦
order to withstand the temperatures well above 2,000◦ F (1093.3 C). The most
advanced of these turbine blades are grown from molten metal as single crys-
tals in ceramic molds, in order to obtain maximum possible resistance to high-
temperature deformation.

2.1.1. High-Performance Aluminum

Aluminum bridge decks have a number of advantages over concrete and steel decks.
Aluminum is about 80 percent lighter than concrete. This substantial weight
savings allows many bridges to be strengthened without extensive reengineering
of substructures.
Aluminum requires fewer welds than steel, eliminating many potential failure
points. Compared to steel, aluminum is less expensive, both in the short term and
the long term. An aluminum deck is also more resistant to corrosion and other
environmental degradation.
Aluminium alloys are often used in Defense and Aerospace, which is one of
the most demanding industries, see Figure 2.2. These industries use high strength
5xxx series (Al-Mg) which are non-heat treatable base alloys for some applications,
but also make use of some of the more specialized heat treatable aluminum alloys
with superior mechanical properties.
Aluminum armor plating is used for its impact strength and strength-to-weight
ratio, and here alloy 5083, 7039 (Al-Zn) and and 2519 (AL-Cu ) are used as base
materials. Missiles are constructed of alloys 2019 and 2219. Perhaps the most
exotic aluminum alloys, with exceptional strength over a wide range of operating
temperatures, are used in the aerospace industry. Some of these alloys are 2219,
2014, 2090, 2024, and 7075. These base materials are typically used in specialized
high performance applications and have their own welding characteristics and
associated problems that require special considerations when joining.

19
 Figure 2.2: Aerospace industry.

2.1.2. High-Performance Steel

A new grade of high performance steel, HPS-485W or HPS-70W, uses a new
chemical composition that provides improved welding and toughness properties.
The increase in strength and performance will allow targeted use of HPS that will
extend the useful life of steel bridge structures; and even greater savings with the
reduction of the total steel weight.
Structural components in new cars are for example made from high-performance
steel that is 3 times stronger than ordinary steel (900 MPa instead of 300 which
is ususal). This makes the new cars very crash resistant such that the passengers
are much more protected if there should be a collision.

2.1.3. Space-age metals

In jet aircrafts, rockets, missiles and nuclear reactors, the metals columbium (Nb,
41), titanium (Ti, 22), hafnium (Hf, 72), zirconium (Zr, 40) and tantalum (Ta,
73) are used. Columbium (niobium) can withstand high temperatures and can be
used for the skin and structural members of aerospace equipment and missiles. It
is a lightweight metal and a superconductor of electricity (which means that it
could be possible to use it in storing and transferring large amounts of energy in
the future). Titanium is as strong as steel and 45% lighter, and is often used for
jet engine components (rotors, fins, and compressor parts) and other aerospace

20
parts. Hafnium and zirconium are always found together are they make it possible
to control nuclear reactors in a precise way. The superior corrosion resistance of
zirconium makes it (and its alloys) very useful in the chemical industry and for
surgical implants. Pure zirconium has for a long time been used as the light
source in photo flash tubes, since it is a reactive metal that burns in air with a
brilliant white light. Tantalum (often occurs as the mineral columbitetantalite) is
◦
very malleable and ductile. Its melting point is at 2996 C, and is often used as a
replacement for platinium in chemical, and dental equipment and instruments. It
is also used to make electrolytic capacitors and are used in vacuum furnaces.
A superconducting material transmits electricity with virtually no energy loss.
Superconductivity, which occurs in many metals and alloys, is very new and is
therefore not yet in widespread use. Superconductivity is a phenomenon occurring
in certain materials at extremely low temperatures, characterized by exactly zero
electrical resistance and the exclusion of the interior magnetic field (the Meiss-
ner effect). Superconductivity occurs in a wide variety of materials, including
simple elements like tin and aluminium, various metallic alloys and some heavily-
doped semiconductors. Superconductivity does not occur in noble metals (metals
that are resistant to corrosion or oxidation, for instance gold, silver, tantalum,
platinum, palladium and rhodium; unlike most base metals), nor in most ferro-
magnetic metals.

Questions:
•Why are metals and metal alloys used?
•What is so special about high performance metals?
• In the space-age, what features in a material or material structure would be
important in the future?
•What kind of metals are used in space equipment, and why are they used?
References:
http://www.mse.cornell.edu/courses/engri111/metal.htm
http://www.allvac.com/
http://www.greatachievements.org/
http://www.hiper-group.com/hcproducts.htm
http://www.spacedaily.com/news/materials-02zh.html
http://www.alcotec.com/ataafi.htm
http://www.weldreality.com/aluminumalloys.htm
High Performance Steel Designers’ Guide:
http://www.fhwa.dot.gov/bridge/guidetoc.htm

21
http://www.mmc.co.jp/alloy/english/products/taisyoku/gijyutsu3.html
R. Gregg Bruce, Mileta M. Tomovic, John E. Neely and Richard R. Kibble, Mod-
ern materials and manufacturing processes, ISBN: 0-13-186859-4, Second edition,
Prentice Hall, 1998.

22
 Figure 3.1: Polymerization by addition of equal monomers.

3. Polymers and Plastics

Polymer etymology: The word polymer comes from Greek: poly means ‘many’
and mer comes from merous which roughly means ‘parts’. Polymers are organic
materials characterized by long chain-like molecules built up from many units
(monomers), see Figure 3.1, generally repeated hundreds or thousands of times.
All atoms in a chain are bonded by covalent bond to each other, while Van der
Waals bonding keeps the chains together. Starch, cellulose, and proteins are
natural polymers. Nylon and polyethylene are synthetic polymers.
From organic chemistry, polymer is defined as: ”a large molecule formed by
the union of at least five identical monomers; it may be natural, such as cellulose
or DNA, or synthetic, such as nylon or polyethylene; polymers usually contain
many more than five monomers, and some may contain hundreds or thousands of
monomers in each chain.”

3.1. The structure of polymers

Polymerization is the word for the process of forming large molecules from small
molecules. When there are no possibility to add any more atoms, the molecules
are said to be saturated. When molecules do not have the maximum number
of atoms, because atoms in the molecule are held together with double or triple
covalent (shared) bonds, they are called unsaturated. The unsaturated molecules
are important in the polymerization process, which means that small molecules
are linked together to form large molecules, as you can see in Figure 3.1, where
the process has taken place through addition mechanism. Here, a large molecule

23
 Figure 3.2: Methane and ethane.

(polymer) is formed by a repeated unit (mer). Activators or catalysts (benzoyl

peroxide) are often required to drive the reaction. A polymer can also be made
from a condensation process which means that reactive molecules combine with
one another to produce a polymer plus small, by-product molecules, such as water.
Often, heat, pressure or a catalyst are required to drive the reaction. In most
commercial available plastics the number of mers in the polymer (which is known
as the degree of polymerization) range from 75 to 750. When two different types of
mers are combined into the same addition chain, we call them copolymers. And,
in terpolymers three different monomers are included.
The molecular structure of polymers are often based on paraffin-type hydro-
carbons where carbon and hydrogen are linked in the relationship

Cn H2n+2 ,

as we see in Figure 3.2. Hydrogen can be replaced by chlorine, flouring and also
benzene. Carbon can be replaced by oxygen, silicon, sulfur or nitrogen. These
possibilities are the reason why a wide range of organic compounds can be created.
By the description of polymers we see that wood (cellulose-type materials, see
Figure 3.3) are included in the group of polymer. It has been accepted for many
years that cellulose is a long chain polymer, made up of repeating units of glucose,
a simple sugar. As a carbohydrate, the chemistry of cellulose is primarily the
chemistry of alcohols; and it forms many of the common derivatives of alcohols,
such as esters, ethers, etc. These derivatives form the basis for much of the
industrial technology of cellulose in use today. Cellulose derivatives are used
commercially in two ways, as transient intermediates or as permanent products.

24
 Figure 3.3: The structure of cellulose.

3.2. Crystallinity in polymers

We need to distinguish between crystalline (see Figure 1.1) and amorphous ma-
terials (see Figure 1.2) and then show how these forms coexist in polymers. The
reasons for the differing behaviors lie mainly in the structure of the solids.
The morphology of most polymers is semi-crystalline. That is, they form mix-
tures of small crystals and amorphous material, see Figure 3.4 and melt over a
range of temperature instead of at a single melting point. The crystalline mate-
rial shows a high degree of order formed by folding and stacking of the polymer
chains. The amorphous or glass-like structure shows no long range order, and the

Figure 3.4: A polymer with the combination of amorphous and crystalline areas, from:
http://plc.cwru.edu/tutorial/enhanced/files/polymers/orient/orient.htm

25
chains are tangled. There are some polymers that are completely amorphous, but
most are a combination with the tangled and disordered regions surrounding the
crystalline areas. Most thermoplastics have crystalline regions alternating with
amorphous regions, while

3.2.1. The glass transition temperature and melting temperature

The glass transition temperature Tg (also called the glass temperature), describes
the temperature at which amorphous polymers undergo a second order phase tran-
sition from a rubbery, viscous amorphous solid to a brittle, glassy amorphous solid.
Tg is a property of amorphous polymers which do not have a sharp melting point,
and is defined as the temperature at which the specific volume vs temperature plot
has a change in slope, see Figure 3.5. When we cool an amorphous material from
the liquid state, there is no abrupt change in volume such as occurs in the case
of cooling of a crystalline material through its freezing point, Tf (=Tm ). Instead,
at the glass transition temperature, Tg , there is a change in slope of the curve of
specific volume vs. temperature, moving from a low value in the glassy state to a
higher value in the rubbery state over a range of temperatures. This comparison
between a crystalline material (1) and an amorphous material (2) is illustrated in
Figure 3.5. Note that the intersections of the two straight line segments of curve
(2) defines the quantity Tg . When a polymer is cooled below Tg , the molecules
have little relative mobility, and the material becomes hard and brittle, like glass.
Some polymers are used above their glass transition temperatures, and some are
used below. Tg is usually referred to wholly or partially amorphous phases in
glasses and polymers. As the temperature of a polymer drops below Tg , it be-
haves in an increasingly brittle manner. As the temperature rises above the Tg ,
the polymer becomes more rubber-like. In general, values of Tg well below room
temperature define the domain of elastomers and values above room temperature
define rigid, structural polymers.
The glass transition is not the same thing as melting. Among synthetic poly-
mers, (crystalline) melting temperature Tm crystalline melting is only discussed
with regards to thermoplastics, as thermosetting polymers will decompose at high
temperatures rather than melt. Tm (also called flow temperature for amorphous
materials) happens when the polymer chains fall out of their crystal structures,
and become a disordered liquid.
Even crystalline polymers will have a some amorphous portion. This portion
usually makes up 40-70% of the polymer sample. This is why the same sample of a

26
Figure 3.5: Comparison between a crystalline material (1) and an amorphous material
(2). From: http://plc.cwru.edu/tutorial/enhanced/files/polymers/therm/therm.htm

polymer can have both a glass transition temperature and a melting temperature.
But you should know that the amorphous portion undergoes the glass transition
only, and the crystalline portion undergoes melting only.
According to Wikipedia, the disaster of the space Shuttle Challenger was
caused by rubber O-rings that were below their Tg , on an unusually cold Florida
morning, and thus could not flex enough to form proper seals between sections
of the two solid-fuel rocket boosters (SRB), see Wikipedia. SRB’s are used to
provide the main thrust (reaction force) in spacecrafts launches from the earthe
up to about 45 kilometres.
Polymer science is a broad field that includes many types of materials which
incorporate long chain structure of many repeat units as discussed above. The
two major polymer classes are:
• Elastomers
• Plastics

Another important property of polymers, also strongly dependent on their

temperatures, is their response to the application of a force, as indicated by two

27
 Figure 3.6: Plastics.

main types of behavior: elastic and plastic. Elastic materials will return to their
original shape once the force is removed. Plastic materials will not regain their
shape. In plastic materials, flow is occurring, much like a highly viscous liquid.
Most materials demonstrate a combination of elastic and plastic behavior, showing
plastic behavior after the elastic limit has been exceeded.

3.3. Plastics
Plastics, see Figure 3.6, are a large group of polymers that has properties between
elastomers and fibers, and has plastic behavior. As such, plastics have a wide
range of properties such as flexibility and hardness and can be synthesized to
have almost any combination of desired properties.
Plastics are polymers which, under appropriate conditions of temperature and
pressure, can be molded or shaped (such as blowing to form a film). In contrast
to elastomers, plastics have a greater stiffness and lack reversible elasticity. All
plastics are polymers but not all polymers are plastics.
Cellulose is an example of a polymeric material which must be substantially
modified before processing with the usual methods used for plastics. Every day
plastics such as polyethylene and poly(vinyl chloride) have replaced traditional
materials like paper and copper for a wide variety of applications.

3.4. Properties of plastics

New types of plastics are developed all the time, and therefore it is helpful to have
a brief knowledge of the following general basic properties of plastics.

28
 • Light weight: Most plastics have a specific gravity (SG) between 1.1 and
1.6. Magnesium has a SG about 1.75. Specific gravity is the heaviness of
a substance compared to that of water, and it is expressed without units. In
the metric system specific gravity is the same as in the English system. If
something is 7.85 times as heavy as an equal volume of water (such as iron
is) its specific gravity is 7.85. Its density is 7.85 grams per cubic centimeter,
or 7.85 kilograms per liter, or 7.85 metric tons per cubic meter.

• Corrosion resistance: Many plastics perform well in hostile, corrosive

environments.
• Electrical resistance: Plastics are often used as insulating materials.

• Low thermal conductivity: Plastics are relatively good thermal insula-

tors.

• Variety of optical properties: You can get a plastic in almost any color
you would like, and the color can go throughout (not only at the surface).
You can also get transparent or opaque plastics.
• Formability: Plastics are easy to form (often in only one single operation).
Extrusion, casting and molding are widely used.
• Surface finish: You can get the surface you want, from rough to excellent
surface finish.

• Comparatively low cost: Both material and processing/manufacturing

processes are cheap. Tool costs are also low.
• Low energy content: Plastics melt at low temperature compared with
metals, and are therefore produced with little energy.

The mechanical strength of plastics are not especially high, compared to met-
als, but the low density makes them comparable to metals due to their strength-
to-weight ratio (or specific strength). A material has high specific strength if the
ratio of its strength to its weight is high.
Plastics are usually divided into two main groups: thermosettings or thermo-
plastics. The terms refer to the materials response to elevated temperature. It is
important to know whether the plastic is a thermosetting or a thermoplastic since
it determines how the plastic will perform in service.

29
3.4.1. Thermosettings
In some plastics, polymerization produces cross linkage between long molecular
weight chain molecules. These plastics are known as thermosetting plastics be-
cause they are permanently hardened by heat. The setting process is irreversible,
so that these materials do not become soft under high temperatures. Additional
heating do not lead to softening, but the material maintain their mechanical prop-
erties up to the temperature at which they char, or burn.
Thermosetting plastics usually have a highly cross-linked or three dimensional
framework structure in which all atoms are connected by strong, covalent bonds.
They are generally produced by the process of condensation polymerization where
elevated temperature promotes the irreversible reaction, hence the term thermoset-
ting.
The thermosettings are stronger and more rigid than the thermoplastics, but
have a lower ductility and poorer impact properties. These plastics also resist
wear and attack by chemicals and they are very durable, even when exposed to
extreme environments.
Typical thermosetting-type plastics are the aminos, most polyesters, alkyds,
epoxies, phenolics and urethanes.

3.4.2. Types of Thermosettings

PUR - Polyurethane/polyurethene (or PU)
PUR is found in a variety of forms ranging from stiff to soft types. Stiff PUR is
used for casings and modelling material (artificial wood). Softer rubber-like PUR
is used for handtools. Expanded PUR is used for mattresses and car inner panels
where it both forms the foam and the leather-like skin. Expandable insulation
foam and moisture hardening glue are also made from PUR. Large parts are often
made from PUR due to low tooling cost. In general it has excellent outdoor
performance and resistance to most acids and solvents.
Genoa stacking chair. PUR can be made from a wide variety of raw materials
to give hard, clear resins for surface coating: soft flexible resins for oil resistant and
abrasion resistant rubbers; rigid or flexible foams for thermal insulation, cushion
and fabric stiffeners. In building surface coating and thermal insulation are their
main applications.
Products: Mil-Tek high pressure cleaner, Panton chair, Shoe soles, and it is
used as modelling material (produced by Westnofa) in the Product Design course
at HiN .

30
EP - Epoxy
Epoxy is a strong and very resistant thermoset plastic. It is used as an adhesive
agent, as filling material, for moulding dies, and as a protective coating on steel
and concrete. Many composite materials are reinforced epoxy.
Epoxy is resistant to almost all acids and solvents, but not to strong bases or
solvents with chlorine content.
By adding a hardening agent curing takes place. The type of hardener has a
major influence on properties and applications of epoxies.
Products: Surfing board, Composite bicycle, Badminton racket, Wheel chair,
Knee support, Soda stream pressure, container, Skull, model, Car space frame,
Wheel chair ramp, Swing wheel.
UP - Unsaturated Polyester
UP is widely used as filler material with glass fibre in sailing boats, hard tops
for cars, furniture, etc. In general UP is not resistant to solvents and bases.
Apart from sulphuric acid it resists acids. To improve appearance and resistance
a surface layer (gel coat or paint) is often added. UP does not require expensive
equipment or tooling to work with UP, and it is therefore often used for prototypes
and low-volume production.
Products: Sailing boat, Chair, Hard top for car, Printed circuit board, Pedes-
trian bridge.
UF - Urea formaldehyde
Urea thermoset molding compounds offer a wide range of applications for
every-day living and industry. Urea formaldehyde (UF) thermosets are economi-
cally priced, they are strong, glossy, and durable. They are not affected by fats,
oils esters, ether, petrol, alcohol or acetone, nor by detergents or weak acids, and
they exhibit good resistance to weak alkalis.
Their high mechanical strength, heat and fire resistance, and good electrical
arc and tracking resistance make them an ideal plastic for numerous industrial and
household applications, from doorknobs and toilet seats to electrical components
and cosmetics enclosures. You name it — if it can be plastic, it can be stronger
and brighter as a (Perstorp) urea thermoset.
MF - Melamine formaldehyde
Melamine thermoset plastics are similar to urea molding compounds, but
melamine has even better resistance to heat, chemicals, moisture, electricity and
scratching.

31
 Melamine formaldehyde (MF) thermosets are ideal for dinnerware, kitchen
utensils, bathroom accessories, and electrical components. The molded com-
pounds are bright, inviting, and highly resistant to scratches and staining. (Per-
storp’s) melamine thermosets are approved for contact with foodstuffs, and they
do not affect the food’s flavor - even at high temperatures. They are very, very
durable.
Like urea molding compounds, melamine thermosets consist of plastic that has
high surface hardness and gloss, brilliant and precise colors, and light fastness.
UF or MF thermosets can be manufactured in a precise and vibrant array of
colors. Two-tone compression molding using doublepunch tools will enable you
to express your creative designs. For example, your cups or sinks can have white
inside and tasteful color outside. The choice of color and shape is limited only by
your imagination.
Alkyd resins
This group of polyesters are usually compression moulded from powders. Orig-
inally produced for inclusions in paints, they are resistant to heat and electricity
and will withstand attack by acids and solvents. Alkyd plastics find applications
in enamels for cars, refrigerators and washing machines and are also used for
electric motor insulation and some television parts.
Silicones
Silicon is an element whose atoms have similar linking properties to those of
carbon, but it is stable at much higher temperatures. To utilize the properties of
silicon, certain plastics, called silicones have been developed based on the element.
Although they are much more expensive, they contain excellent properties. These
super materials are available as oil, plastics and rubbers. To quote an example
illustrating their value: silicone rubber will retain its elasticity over a temperature
range of -80◦ C to 250◦ C. The stability of silicones under widely varying service
conditions make them valuable materials for applications such as laminates in the
aerospace industry, gaskets and seals for engineering purposes and cable insulation
for aircraft electrical systems.

3.4.3. Thermoplastics
Thermoplastics, alternatively thermosoftenings, as the name implies, are hard at
low temperatures but soften when they are heated. The softening and hardening
can be repeated without any change in the chemical structure. Although they are

32
Figure 3.7: The easiest way to identify the type of thermoplastic you’re working with
is to look for the Plastic ID symbol the backside of the part.

less commonly used than thermosetting plastics they do have some advantages,
such as greater fracture toughness, long shelf life of the raw material, capacity for
recycling and a cleaner, safer workplace because organic solvents are not needed
for the hardening process.
Thermoplastics have weak bonds between the neighboring molecules and they
are weakened by elevated temperature which means they soften at high temper-
ature and are stronger and harder when cooled. They do not have any definite
melting temperature, they have a range of temperatures where they soften. When
cooled below the glass transition temperature, Tg , the linear polymer retains its
amorphous structure, but becomes hard, brittle, and glasslike.
Thermoplastics are not cross-linked and can be softened and hardened over and
over again. The majority of polymers are thermoplastic. Today there are primar-
ily six commodity polymers in use, namely polyethylene terephthalate (PETE),
polyethylene (PE), polyvinyl chloride (PVC), polypropylene (PP), polystyrene
(PS) and polycarbonate (PC). These make up nearly 98% of all polymers and
plastics encountered in daily life, see ref.[33].
The Society of the Plastics Industry, Inc. (SPI) introduced its resin identifica-
tion coding system in 1988 at the urging of recyclers around the country. The SPI
code was developed to meet recyclers needs while providing manufacturers a con-
sistent, uniform system that could apply nationwide, see figure 3.7. Look at the
bottom of a recyclable plastic bottle - chances are you will see a PE or PS which
means polyethylene or polystyrene. These materials are examples of what happens

33
to polymers when they solidify: the chains are entangled and packed together to
make light, tough, flexible materials. If you heat up PE or PS to moderate tem-
peratures, if the chains have not been chemically stuck together (‘cross-linked’)
they will melt, and turn into goopy liquids, which are called polymer melts. Some
polymers melts even at room temperature, like polydimethylsiloxane (PDMS), or
poly(ethylene-propylene) (PEP).

3.4.4. Types of Thermoplastics

PET - Polyethylene terephthalate (or PETE)
PET has good barrier properties against oxygen and carbon dioxide. There-
fore, it is utilized in bottles for mineral water. Other applications include food
trays for oven use, roasting bags, audio/video tapes as well as mechanical compo-
nents.
PET exists both as an amorphous (transparent) and as a semi-crystalline
(opaque and white) thermoplastic material. Generally, it has good resistance
to mineral oils, solvents and acids but not to bases.
The semi-crystalline PET has good strength, ductility, stiffness and hardness.
The amorphous PET has better ductility but less stiffness and hardness.
Danish Name PET - thermoplastic polyester
Products: Bottle for mineral water, Trays for oven use, Oven foils, Audio and
video tapes, Thermo scarf (fleece)
PE - Polyethylene
PE is a semi-crystalline thermoplastic material and one of the most commonly
used plastics. It is generally ductile, flexible and has low strength. PE is one
of the most commonly used thermoplastic material due to the good properties
combined with a low price.
There are two basic families: LDPE (low density), and HDPE (high density):

• LDPE - low density polyethylene: LDPE is the low density version of

PE. This has less hardness, stiffness and strength compared to HDPE, but
better ductility. It is opaque and only thin foils can be transparent. LDPE
is used for packaging like foils, trays and plastic bags both for food and
non-food purposes. Used as protective coating on paper, textiles and other
plastics, for instance in milk cartons. Products: Wrapping foil for packaging,
Plastic bag (soft type that does not crackle), Garbage bag, Tubes, Ice cube
plastic bag.

34
 • HDPE - high density polyethylene: HDPE is the high density version
of PE plastic. It is harder, stronger and a little heavier than LDPE, but less
ductile. Dishwasher safe. HDPE is lighter than water, and can be moulded,
machined, and joined together using welding (difficult to glue). The appear-
ance is wax-like, lusterless and opaque. The use of UV-stabilizators (carbon
black) improves its weather resistance but turns it black. Some types can be
used in contact with food. Products: Milestone, Bottle for motor oil, Bot-
tle for organic solvents, Street bollard, Hedge cutter, Gasoline tank, Milk
bottles, Plastic bag (stiff type that crackles), Children’s toys, Lid for honey
pot, Beer crate, Dolphin bicycle trailer.

PVC - Polyvinyl chloride (vinyl)

PVC is one of the oldest and most commonly used thermoplastic material, it
is a heavy, stiff, ductile and medium strong amorphous (transparent) material.
By adding softeners, a range of softer materials can be achieved, ranging from
a flexible to an almost rubber-like elastic soft material. Softeners also help to
increase the manufacturability. PVC has brilliant resistance to acids and bases,
but is affected by some solvents. Soft PVC is exceptionally resistant to most
chemicals. The poor weather resistance can be improved using additives. PVC
has good barrier properties to atmospheric gasses. PVC has a Tg of 83◦ C, making
it good, for example, for cold water pipes, but unsuitable for hot water. PVC will
also always be a brittle solid at room temperature.
Products: Boat fender, garden hose, electrical wire insulation, vinyl flooring,
roof gutter, vinyl record, children’s doll, wrapping film, medical transparent tube,
Pneumatic chair.
PP - Polypropylene
PP is an inexpensive, ductile, low strength material with reasonable outdoor
performance. The material surface is soft wax-like and scratches easily. It has a
high stiffness, good strength even in relatively high temperatures, abrasion resis-
tant, good elastic properties and a hard glossy surface. In low temperatures PP
gets brittle (< 0◦ C). Stiffness and strength are often improved using reinforcement
of glass, chalk or talc. The color is opaque and white, but it can be dyed in many
colors. In many ways, PP is similar to HDPE, but it is stiffer and melts at 165-
170◦ C. PP can be manufactured by all the methods used for thermoplastics. PP
has high crystallinity (70-80 %), and is one of the lightest thermoplastics on the
marked. The chemical properties are good. PP is resistant to inorganic chemicals

35
and water. It is resistant to most strong mineral acids and basics. PP is not
resistant to nitrous gasses, halogens and strong oxidizing acids.
Products: Childrens toy bin, transport box, fuel tank, suitcase, garbage bin,
rope, shaver (rechargeable), air intake, tubes, packing material, auto parts etc.
The material is often used for hinges as it can be flexed millions of times before
breaking.
PS - Polystyrene
Polystyrene is an inexpensive amorphous thermoplastics that has good me-
chanical proprieties. It is vitreous, brittle and has low strength. However it is
also hard and stiff. Foamed PS is used for packaging and insulation purposes.
PS is not weather resistant, and therefore not suitable for outdoor uses. PS is
transparent (it transmits about 90% of the sunlight) and has unlimited dyeing
possibilities. Assembly can be done with gluing.
Products: CD and MC covers, disposable drinking glass, glass for bicycle
lamp, salad bowl, razor (ordinary), razor (biodegradable), disposable articles,
signs, machine parts and picture frames etc.
PC - Polycarbonate
Polycarbonate is an amorphous plastic with very high impact strength, good
ductility and high stiffness. It is very difficult to break and the material is therefore
considered fracture-proof (e.g. bullet-proof glass).
Light transmission is 85-90% but depends on the thickness. It has good out-
doors resistance in the UV-stabilized form, but it tends to turn yellow by long
exposition to sunlight. PC is transparent and can be dyed in many colors. PC
has a relatively good chemical resistance.
Products: PC is commonly used for shielding of work places and machines,
sight glass, tubes etc. due to its transparency and high impact resistance, CD
compact disc, bullet-proof glass, water container.
ABS -Acrylonitrile-butadiene- styrene
Acrylnitrile contributes with thermal and chemical resistance, and the rub-
berlike butadiene gives ductility and impact strength. Styrene gives the glossy
surface and makes the material easily machinable and less expensive.
Generally, ABS has good impact strength also at low temperatures. It has sat-
isfactory stiffness and dimensional stability, glossy surface and is easy to machine.
If UV-stabilizators are added, ABS is suitable for outdoor applications.
Products: LEGO building bricks, Computer mouse, Vacuum jug, KimBox
suitcase, Ceramic advanced wet shave razor, Hedge cutter handle, Handle for

36
high pressure cleaner, Shaver, rechargeable, Ensemble chair (ABS blended with
PA), auto body parts, suitcases, toys etc. Extruded profiles, tubes and bolts can
be made from ABS when the requirements are high impact resistance and a nice
surface.
PA - Polyamide (nylon)
PA is a group of amorphous (transparent) and semi-crystalline (opal-white)
plastics. Arguments for using PA include strength (fishing line, axe handle),
wear resistance (bearings), barrier properties (food packaging) and machinability.
Polamide is recognised for good abrasion resistance, low friction coefficient, good
resistance to heat and good impact resistance. PA absorbs water which makes it
softer. UV-stabilizators are required for outdoor applications.
Products: Nylons (stockings), fishing line, bicycle trailer (rainproof cover),
bearing, axe, hedge cutter, handle for, high pressure cleaner, bottle for tomato
ketchup (barrier layer), ensemble chair (PA blended with ABS), bottle-opener.
Kevlar (aramid fibre) is a family of nylons.
Acrylic - PMMA (plexiglas)
PMMA (polymethyl-methacrylate) is an amorphous thermoplastic material
with very good optical properties (as transparent as glass and it allows 92% of
the sunlight to pass!).
PMMA is hard, stiff and medium strong, easy to scratch, notch sensitive,
but easy to polish and has a very good weather resistance. Exceptional outdoor
performance, such as weather and sunlight resistance, without reduction neither of
optical nor mechanical properties. PMMA is resistant to water, basics, inorganic
salts diluted in water, most diluted acids. It is not resistant to strong acids, basics
and polar solvents.
Products: Tail light glass, exhibition case, folding chair, kitchen scale, decora-
tion articles, transparent tubes, signs, windows, level glass etc..
POM - (polyoxymethylene) Acetal
Acetal is a crystalline plastic often used for technical applications due to its
strength, ductility and good machinability. It has good creep properties which
makes it suitable for click connections (e.g. bicycle lamp holder). It exhibits good
stiffness, strength and hardness up to 120◦ C, and the elasticity is comparable
with that of many metals. Acetal is wear resistant and has a very low friction
coefficient. The color is opaque white. To protect it from UV-light carbon black
can be added, changing its color into black.

37
 Products: Vacuum jug (top), Holder for bicycle lamp, Fitting for Vico Duo
chair, Gearwheel.
PTFE - Fluoropolymer (Teflon)
Fluoropolymers can be used to make a variety of articles having a combination
of mechanical, electrical, chemical, temperature and friction-resisting properties
unmatched by articles made of any other material. Commercial use of these and
other valuable properties combined in one material has established TEFLON° R

resins as outstanding engineering materials for use in many industrial and mil-
itary applications. TEFLON° R
resins may also be compounded with fillers or
reinforcing agents to modify their performance in use.
TEFLON° R
PTFE resins have a continuous service temperature of 260◦ C
◦
(500 F). Much higher temperatures can be satisfactorily sustained for shorter
exposures.
Teflon is used in: Food processing, Electrical parts, Coaxial cable connectors,
Terminal insulators, Transformers, Relays, Medical industry, Washers, Gaskets,
Flanges, Valve components, Pump Components, Baffles, Seals, Bearings, Rings,
Bushings , High heat applications. Teflon is available as: Sheets, Rods, Tubes,
Heavy wall tubing, Film, Rectangular bar, Pressure sensitive, tape.
TPUR - Thermoplastic urethane (or TPU)
TPU is a urethane based TPE (thermoplastic elastomer). It is made of long
chained molecules of diols and diisocyanates. Urethane is the unit which is re-
peated in polyurethane. TPU has a very good abrasion resistance. It is a tough
material with a good elasticity over a wide temperature range. TPU is filling the
gap between rubber material and the more traditional thermoplastics.
The electrical conductance is very low. TPU is a hygroscopic material, and
the conductance is dependent on the content of moisture. TPU is resistant to
oil, fat, gasoline, and ozone. It is not resistant to hot water, steam, strong acids
and basics. Polyether based TPU is resistant to microbes. TPU is available in a
variety of qualities.
Products: Bumpers, hoses, tubes, sleeves, cushions, coating, insulators, apron
rollers etc.
PEEK - Polyetheretherketone
PEEK is a high temperature resistant engineered thermoplastic with excel-
lent chemical and fatigue resistance plus thermal stability. They exhibit superior
mechanical and electrical properties. With a maximum continuous working tem-
perature of 249◦ C (480◦ F), they have excellent retention of mechanical properties

38
up to 299◦ C (570◦ F) in a steam or high-pressure water environment. Superior
chemical resistance has allowed them to work effectively as a metal replacement
in harsh environments. They are inert to all common solvents and resist a wide
range of organic and inorganic liquids. When ∆extensive machining is required,
a secondary annealing process should be considered.
PEEK is an excellent material for a wide spectrum of applications where ther-
mal,chemical, and combustion properties are critical to performance. The addi-
tion of glass fiber and carbon fiber reinforcements enhances the mechanical and
thermal properties of the basic PEEK material.
Products: automobile engine parts, medical equipment, aerospace products.

Today there are primarily six commodity polymers in use, namely polyethylene,
polypropylene, polyvinyl chloride, polyethylene terephthalate, polystyrene and
polycarbonate. These make up nearly 98% of all polymers and plastics encoun-
tered in daily life.

3.5. Classification
• Standard plastics used in non-critical and low-stress applications are mate-
rials like PS, ABS, PVC, PP, HDPE, and LDPE.

• Engineering plastics that are used in general structural, bearing and wear
purpose are plastics like PPO (Polyphenylene oxide, modified), Acrylic, PC
(Polycarbonate), PET-P (Polyethylene Terephtalate), POM (Poly-oxymethylene
=Acetal), PA (Polyamide = Nylon), and UHMW-PE (Ultra high mole wt.
Polyethylene).

• Advanced engineering plastics that have superior properties and can be

used in extreme environments are plastics like PSU (Polysulfone), PPSU
(Polyphenylsulfone), PEI (Polyetherimide), PTFE (Polytetraflouroethylene
= Teflon), PPS (Polyphenylene sulfide), PEEK (Polyetheretherketone), PI
(Polyimid), PAI (Polyamide-imide), and PBI (Polybenzimidazole).

3.6. Additives in Plastics

Very often, some additional materials are mixed into plastics, to obtain:

• improved properties

39
 • reduced cost

• improved moldability

• wanted color

These additional materials are classified as fillers, plasticizers, lubricants, color-

ing agents, stabilizers, antioxidants, flame retardants, foaming or blowing agents,
antifogging agents, antistatic agents, clarifying agents and optical brighteners.
New ones are added to the list continuously, so we briefly mention a few of them.
Fillers
Improve mechanical properties, reduce shrinkage, reduce weight or provide
bulk. Fillers comprise a large percentage of the total volume of the plastic. Ex-
ample of fillers can be: wood flour, cloth fibers, glass fibers, clay.
Plasticizers
Increase flexibility, improve flow during molding, reduce shrinkage, reduce
weight.
Lubricants
Comparison of both internal lubricants and external lubricants which can be
blended with various materials to reduce friction, and wear, improve mar resis-
tance, and extend the useful life of products which are subject to friction. They
also improve moldability and extraction from molds.
Coloring Agents
Coloring Agents are put in the plastic to Impart color.
Stabilizers
Stabilizers retard degradation due to heat or light.
Antioxidants
Antioxidants retard degradation due to oxidation.
Flame retardants
Flame retardants reduce flammability.
Foaming Agents
Foaming Agents, also known as Blowing or Nucleating Agents, can eliminate
sink marks, reduce density, shorten cycle time and reduce total production costs.
In extrusion and injection molding, foaming agents can save material weight and
lower total cost. They also improve extrusion rates by increasing the volume that

40
can be processed per extruder in a given period of time, and endothermic foaming
agents absorb heat and improve injection molding cycle time.
Antifogging agents
Antifogging agents reduce the formation of condensed droplets on the surface
of polyolefin films, such as for food-packaging or agricultural films, resulting in
better film transparency and consequently providing better food preservation.

Antistatic agents
These new permanent antistatic agents, form a conductive network throughout
the polymer matrix which dissipates the electrical charge as it builds up. They
are already effective for processing and even work at low environmental humidity.
Clarifying agents/Optical brighteners
Clarifying agents/optical brighteners are designed to give brilliance and whiten-
ing to a variety of applications. Synthetic fibers for example, have an inherent
yellowish tint. Add optical brighteners and the fibers appear cleaner and whiter.
IRGACLEAR° R
is a range of products that not only improves the clarity and
transparency of polypropylene, but also enhances the mechanical properties.

3.6.1. Oriented Plastics

The strength of the intermolecular bond increases with reduced separation dis-
tance, and that is why a processing that makes the molecules align parallel make
the long-chain thermoplastic higher strength in a given direction. This is called
an orientation process, and can be obtained by stretching, rolling, or extrusion, as
shown in the Figure 3.8. When a polymer (amorphous or crystalline) is subjected
to stress (tensile), the molecular chains become aligned or oriented parallel to the
direction of applied stress. The polymer is then in an oriented state. Orienting
may increase the tensile strength by more than 200%, but 25% is more typical.

3.7. Elastomers & Rubbers

Elastomers, or rubbery materials, have a loose cross-linked structure. Natural and
synthetic rubbers are both common examples of elastomers. Elastomers possess
memory, that is, they return to their original shape after a stress is applied.
Elastomers are amorphous polymers and consist of long polymer chains above
their glass transition temperature. The structure of elastomers is tightly twisted
or curled.

41
Figure 3.8: Presentation of the alignment of the plastic molecules in the orienting
process.

Figure 3.9: Elastomers.

Elastomers are reversibly stretchable for small deformations. When stretched,

the polymer chains become elongated and ordered along the deformation direction.
When no longer stretched, the chains randomize again. The cross-links guide the
elastomer back to its original shape. They are very flexible and elastic, which
means that they can undergo large elastic deformations without ruptures and
recover substantially in shape and size after the load has been removed. Many
elastomers can be stretched to several times their original length. Also, the cycle
can be repeated numerous times with identical results, as with the stretching of
a rubber band.
Elastomers are generally resistant to oil and fuel, impermeable to liquids and
gases, but tend to deteriorate by oxidation.

42
 Like plastics, elastomers are either thermoplastic material (they can be remelted)
or thermoset material (that cannot be remelted). Rubber is an older name for
elastomers.
Elastomeric polymers do not follow Hooks’s law (as most engineering material
do). The behavior of the elastomers is a bit more complex due to the molecular
shape and the fact that small degree of viscous deformation in produced when
load is applied.

3.7.1. Rubber and Artificial Elastomers

The oldest commercial elastomer is natural rubber, which is made from a processes
sap of a tropical tree. Natural rubber (NR) is a biopolymer which is known as
polyisoprene. The rubber tree (Hevea brasiliensis) is the most common source of
natural rubber used today. Polyisoprene can also be synthesized by polymerization
from its monomer isoprene (CH2=C(CH3)CH=CH2), (IR). This is a rare example
of a natural polymer that we can make almost as well as nature does. Rubber has
been used for centuries by the South American Indians. They most probably were
the people who discovered that if latex (a milky fluid that circulates in the inner
portions of the bark of many tropical and subtropical trees and shrubs) is dried,
it can be pressed into useful objects such as bottles, shoes and balls. Figure 3.10
shows the common use of rubber.
However, it was not until the 1830s (when John Haskins and Edward Chaffee
organized the first rubber-goods factory in the United States) that the commer-
cial rubber industry really began to flourish. But rubber had many weaknesses;
it softened with heat and hardened with cold; it was tacky, odorous, and per-
ishable. In 1834 the German chemist Friedrich Ludersdorf and the American
chemist Nathaniel Hayward discovered that if they added sulfur to gum, then
rubber lessened or eliminated the stickiness of finished rubber goods. Charles
Goodyear discovered in 1839, that cooking natural rubber with sulfur removed
the gum’s unfavorable properties and could be strengthened by cross-linking it
with approximately 30% sulfer and heating it to a suitable temperature. This
process is called vulcanization, and led to many and varied applications of nat-
ural rubber, since the rubber then has got increased strength and elasticity and
greater resistance to changes in temperature. It is also impermeable to gases, and
resistant to abrasion, chemical action, heat, and electricity, in addition to have
high frictional resistance on dry surfaces and low frictional resistance on water-wet
surfaces. The vulcanization process remains fundamentally the same as it was in

43
 Figure 3.10: Common use of rubber.

1839.
Vulcanized rubber has numerous practical applications, such as

• excellent abrasion resistance which makes it valuable for the treads of vehicle
tires, conveyor belts (soft rubber), pump housings and piping used in the
handling of abrasive sludges (hard rubber).

• flexibility characteristics which makes it suitable for use in hoses, tires, and
rollers for a wide variety of devices ranging from domestic clothes wringers
to printing presses.

• its elasticity makes it suitable for various kinds of shock absorbers and for
specialized machinery mountings designed to reduce vibration. Since it its
relatively impermeable to gases its used as air hoses, balloons, balls, and
cushions.

• resistance to water and to the action of most fluid chemicals has led to its
use in rainwear, diving gear, and chemical and medicinal tubing, and as a
lining for storage tanks, processing equipment, and railroad tank cars.

44
 • high electrical resistance, soft rubber goods are used as insulation and for
protective gloves, shoes, and blankets. Hence, hard rubber is used for articles
such as telephone housings, parts for radio sets, meters, and other electrical
instruments.

• coefficient of friction (resistance to movement) of vulcanized rubber, which

is high on dry surfaces and low on wet surfaces, leads to the use of rubber
both for power-transmission belting and for water-lubricated bearings in
deep-well pumps.

• uncertainty of price and supply of natural rubber led to the development

of artificial elastomers. Some artificial elastomers are inferior to natural
rubber, others have superior characteristics.

Rubber that has not undergone the vulcanization process has few practical
uses because of its poor heat resistance and high plasticity. It is used for cements,
adhesive, insulating, friction tapes and for crepe rubber used in insulating blankets
and footwear.

Some other elastomers include:

Polybutadiene (BR): Polybutadiene was one of the first types of synthetic
elastomer, or rubber, to be invented. It is very similar to natural rubber, poly-
isoprene, and is good for uses which require exposure to low temperatures. Tires
treads are often made of polybutadiene copolymers. Belts, shoe soles, hoses, gas-
kets and other automobile parts are made from polyubutadiene, because it stands
up to cold temperatures better than other elastomers.
Polyisobutylene (PIB): Polyisobutylene is a synthetic rubber, or elastomer.
It is the only rubber that is gas impermeable, (can hold air for long periods of
time). For instance, balloons will go flat after a few days because they are made
of polyisoprene, which is not gas impermeable. Since polyisobutylene will hold
air, it is used to make things like the inner liner of tires, and the inner liners of
basketballs.
Polyisobutylene, sometimes called butyl rubber, is a vinyl polymer, and is very
similar to polyethylene and polypropylene in structure, except that every other
carbon is substituted with two methyl groups.
Poly(styrene-butadiene-styrene) (SBS) : Poly(styrene-butadiene-styrene),
is a hard rubber, and is therefore used in soles of shoes, tire treads, and other places

45
 Figure 3.11: Poly(styrene-butadiene-styrene), or SBS.

where durability is important. It is a type of copolymer called a block copolymer.

Its backbone chain is made up of three segments, where the first is a long chain
of polystyrene, the middle a long chain of polybutadiene, and the last segment is
another long section of polystyrene, see Figure 3.11.
Polystyrene is a tough hard plastic, which gives SBS its durability, while
polybutadiene is a rubbery material, and gives SBS its rubber-like properties.
The material the ability to retain its shape after being stretched; SBS is a type of
unusual material called a thermoplastic elastomer.
Polyurethanes (AU or EU): (See also 3.4.2) Polyurethanes are the most
well known polymers that are used to make foams. But, polyurethanes are more
than foam. Polyurethanes are the single most versatile family of polymers there is.
Polyurethanes can be elastomers, and they can be paints. They can be fibers, and
they can be adhesives. You can find them everywhere. A well-known polyurethane
is spandex (DuPont sells it under the trade name Lycra). Lycra is a fiber that
acts like an elastomer, and allows us to make fabric that stretches for exercise
clothing and the like.
Polychloroprene (CR) : Polychloroprene is usually sold under the trade
name Neoprene, and it is especially resistant to oil, and was the first synthetic
elastomer that became a hit commercially. It was Arnold Collins, while work-
ing under the same fellow who invented nylon, Wallace Carothers that invented
polychloroprene.
Silicones (Q): (See also 3.4.2) Silicones can be used for a lot of things. Sili-
cones are inorganic polymers, that is, there are no carbon atoms in the backbone
chain. The backbone is a chain of alternating silicon and oxygen atoms. Each
silicone has two groups attached to it, and these can be any organic groups. Sili-

46
 Figure 3.12: Biopolymers.

cones can stand high temperatures without decomposing, but they have very low
glass transition temperatures.

3.8. Biopolymers
Biopolymers are an alternative to petroleum-based polymers (traditional plastics)
produced by living organisms. The field of biopolymers, is still in its early stage,
but is growing in popularity every day. The term biopolymers is often used for all
polymers that are made from natural renewable resources and/or are completely
biodegradable. Biopolymers can be produced by biological systems like micro-
organisms, plants and animals, or chemically synthesized from biological starting
materials (e.g. sugars, starch, natural fats or oils, etc.). Most of the biopolymers
are biodegradable, some of them are even water soluble, most of the biopolymers
are compostable or will biodegrade in landfill, time can vary from a couple of days
to even years, but they will eventually degrade.
Biopolymers can be used in all kinds of applications, and with all kinds of pro-
duction techniques, like injection moulding, thermoforming and blow moulding!
Some biopolymers can directly replace synthetic plastics in traditional applica-
tions, while others possess unique properties that may open new applications.

3.8.1. What makes a polymer a biopolymer?

Biopolymers are defined as biologically degradable polymers. Biopolymers are
polymers that are generated from renewable natural sources, are often biodegrad-

47
able, and not toxic to produce. According to the American Society for Testing and
Materials (ASTM), biopolymers are degradable polymers in which degradation
results from the action of naturally occurring micro-organisms such as bacteria,
fungi and algae.

3.8.2. Applications of biopolymers

Biopolymers can be used for a lot of applications, from packaging to disposables,
from diapers to cottonsticks, from carparts to bottles, etc. In theory conventional
plastics may be substituted by biopolymers in many applications. In practice sub-
stitution is not always feasible, wanted or the most lucrative way to use biopoly-
mers. Besides technical development which is needed for some applications, the
applications have to be economically feasible within a reasonable term. The eco-
nomic feasibility depends on the investments needed for material and product
development and the added value of the biopolymer in the application.

3.8.3. Properties of biopolymers

Properties of biopolymers depend on the raw material they are based on, on
additives used and on the (chemical) modifications during production.
Different types of biopolymers
Starch, proteins and peptides, and DNA and RNA are all examples of biopoly-
mers, in which the monomer units, respectively, are sugars, amino acids, and nu-
cleic acids. Biopolymers found in and used by living cells can be divided into the
following four classes:

• Lipids (fats, oils, phospholipids waxes, and steroids)

• Carbohydrates (Starch, Cellulose, and Sugars as polysaccharides)

• Proteins (Protein is a polymer with the monomer made up of amino-acid

residues, and is contained in skin, bones, muscles, blood and hair )

• Nucleic acids (complex polymeric molecules which store and translate ge-
netic information )

The most common way to divide biopolymers in different types is on the

basis of the raw material used for production. However, some biopolymers can

48
be produced from different raw materials. Some examples of different types of
biopolymers are:

Starch-based polymers (SBP) are often a blend of starch and other plas-
tics (e.g PE), which allows for enhanced environmental properties. Starch is a
polymeric carbohydrate (a polysaccharide), in which the monomers are glucose
units joined to one another
Starch is abundantly present in many crops. The starch is stored in granules
within the plant. This facilitates the isolation from the plant. Starch may be
modified in order to become a thermoplastic. This makes the starch polymer
suitable for current processes in the plastics industry like: injection moulding
and extrusion. Thermoplastic starch has an affinity with moisture. The material
therefore is not suitable for wet food packaging applications. Direct contact with
water only is possible for a short time. By acetylation a certain resistance to
water may be achieved. By adding PCL the flexibility of the bioplastic increases.
Starch polymer has good oxygen barrier properties. Starch polymers are the most
produced and used biopolymers at the moment. Starch (in particular cornstarch)
is used in cooking for thickening foods such as sauce. In industry, it is used in the
manufacturing of adhesives, paper, textiles and as a mold in the manufacture of
sweets such as wine gums and jelly beans. It is a white powder, and depending
on the source, may be tasteless and odourless.
Cellulose polymers (cellulose esters, cellulose ethers, cellophane)
Cellophane is one of the oldest packaging materials. Cellulose pulp from trees
or cotton may be used to produce cellophane. It is a relatively expensive packaging
material which can be used for a wide range of products such as cd’s, candy and
cigarettes. The higher price is a reason why a large market share was lost to
polypropylene. At the moment new applications in which the specific properties
may be used, are sought for. The material is transparent and has good folding
properties. The foil has a high gas barrier and if a coating is applied may resist
water vapor as well.
Protein polymers
There is no information about protein polymers available yet, but see for in-
stance http://www.ppti.com/Technology/technology.htm.

49
3.8.4. Additional information on raw materials in biopolymers
Lactic acid is produced by the microbial fermentation of sugars such as glu-
cose or hexose. Feedstocks can include potato skins and corn. The lactic acid
monomers can be used to create low or high molecular weight polylactide poly-
mers (PLA). PLA commodity polymers are being developed for use as pulping
additives in paper manufacturing and as biodegradable packing materials, clothes,
cups, packaging and many other everyday products.
Polylactic acid (PLA) is derived from lactic acid for which carbohydrates
in sugar beets, potatoes, wheat, maize and milk are the source. Polylactic acid
is a substance familiar to the human body, as we ourselves produce it by every
muscle contraction. It can be broken down by the body.
PLA can be processed through for example injection moulding, foil blowing
and deep drawing. PLA may be applied as a coating. PLA is water resistant but
cannot withstand high temperatures (>55◦ C). In comparison to starch biopolymer
the degradation process is very slow. However, within a composting facility it can
be broken down in 3 to 4 weeks.
Polyhydroxyalkanoates (PHAs) (polyhydroxybutyrate (PHB), polyhydrox-
ybutyrate/valerate (PHB/HV))
PHA’s are generally derived through fermentation of glucose, sucrose of fatty
acids by micro-organisms.
PHB can be processed through for example injection moulding and deep draw-
ing. It is an excellent material for the coating of paper coffee cups. The most
important features of PHB are the resistance to high temperatures up to 120◦ C
and the resistance to water.
Chitin, a polysaccharide found in the exoskeletons of insects and shellfish,
possesses many desirable characteristics. Chitin’s most important derivative, chi-
tosan, is nearly a ”model” biopolymer with it’s useful physical and chemical prop-
erties, high strength, biodegradability, and nontoxicity. In fact, chitosan brings
new meaning to the word ”biodegradable” as the human body easily breaks it
down into simple carbohydrates, carbon dioxide, and water. This accounts for
the research that is trying to use chitosan in drug delivery systems.
Other natural Biopolymers:
Protein
Proteins are biopolymers consisting of one or more strings of amino acid
residues joined head-to-tail via peptide bonds. Protein also makes up much of

50
the structure of animals: collagen and keratin are components of skin, hair, and
cartilage; and muscles are composed largely of proteins.
Peptides
Peptides are the family of molecules formed from the linking, in a defined
order, of various amino acids. Peptides differ from proteins, which are also long
chains of amino acids, by virtue of their size. Traditionally, those peptide chains
that are short enough to make synthetically from the constituent amino acids are
called peptides rather than proteins. The dividing line is at approximately 50
amino acids in length, since naturally-occurring proteins tend, at their smallest,
to be hundreds of residues long.
DNA
Deoxyribonucleic acid (DNA) is the molecule in living things that contains
the coding information for creating proteins. It is also the molecule of heredity—
whenever an organism reproduces, each offspring gets a copy of its parents’ DNA.

RNA
Ribonucleic acid, a nucleic acid structurally distinguished from DNA by the
presence of an additional hydroxyl group attached to each pentose ring, and func-
tionally distinguished by its multiple roles in the intracellular transmission of
genetic information from the site of transcription (from DNA) to the site of trans-
lation (into protein).

3.8.5. Plastic taste better with sugar

Chemists in India are lacing plastics with sugar to make them palatable to soil
bacteria. The plastics, which normally survive for decades in landfills, start to
biodegrade within days.
The tweaked plastics are polythene, polystyrene and polypropylene. These
make up around a fifth of urban waste by volume. Bottles, bags and sacks are
made of polyethylene, food packaging is made of polypropylene, and drinking
cups, fast-food cartons and the hard casing of electronic equipment are fashioned
from polystyrene.
Digambar Gokhale and colleagues at the National Chemical Laboratory in
Pune mix the styrene subunits of polystyrene with small amounts of another
substance that provides a chemical hook for sucrose or glucose pieces. They then
add sugars to the styrene chains like pendants on a necklace.

51
Figure 3.13: One fifth of urban rubbish is plastics like polystyrene. °
c GettyImages

By weight, less than 3% of the final polymer is sugar, so the material is more
or less the same. But bacteria such as Pseudomonas and Bacillus break open the
chains when they chomp on these sugary snacks, kicking off decay.
It remains to be seen whether the polymer biodegrades into entirely non-toxic
substances. Fully broken down, the end products are carbon dioxide and water.
But along the way, all sorts of other compounds are produced, such as organic
acids and aldehydes.
Indeed, it is not yet clear how far, or how quickly, the plastic will break down
in the real world. And adding the sugar would require significant manufacturing
changes, which could be costly.
Other additives that make polythene, polystyrene and polypropylene biodegrad-
able have been toxic and can leach out of garbage. Another approach is to initiate
the breakdown process using heat, ultraviolet light or exposure to oxygen, but
this is cumbersome and expensive.
Published in: Nature News Service, 2.december 2002; see also
http://www.nature.com/nsu/021125/021125-12.html.
References:
Polymers:
http://www-materials.eng.cam.ac.uk/mpsite/
http://www.plasticsusa.com/
http://www.polymer-age.co.uk/directry/plastic.htm
http://www.mse.cornell.edu/courses/engri111/
http://www.amco.ws/home.asp

52
http://naturalrubber.cjb.net/
http://www3.jaring.my/inro/
http://www.phy.uni-bayreuth.de/theo/tp3/members/elast.htm
http://www.psrc.usm.edu/macrog/crystal.htm
http://www.uvi.edu/Physics/SCI3xxWeb/Structure/Polymerization.html
http://www.zeusinc.com/
http://islnotes.cps.msu.edu/trp/toc.html
http://abalone.cwru.edu/tutorial/enhanced/files/textbook.htm
http://www.thermosets.com/thermosets.html
http://www.aipma.org/t&ti/thermos.htm
http://www.polymer-age.co.uk/techlink.htm
http://en.wikipedia.org/wiki/Glass_transition_temperature
http://en.wikipedia.org/wiki/Polymer#Polymer_science
Elastomers:
http://www.psrc.usm.edu/macrog/elas.htm
Biopolymers:
http://www.biopolymer.net/
http://www.biopolymer.com/
http://www.biotec.de
http://www-classes.usc.edu/engr/ms/125/MDA125/biopolymers/
http://online.itp.ucsb.edu/online/infobio01/lancet2/
http://www.cheresources.com/biopoly2zz.shtml
http://www.proterra.nl/norms.html
http://www.dblab.helsinki.fi/~rtuma/biopolymers.htm
http://www.wikipedia.org/wiki/Biopolymer
http://www.brooklyn.cuny.edu/bc/ahp/SDPS/PSLectNotes/SD.PS.LectP2.html
http://www.bipp.nl/about_3.html
http://www.forskning.no/Artikler/2002/desember/1038997297.18
http://www.nature.com/nsu/021125/021125-12.html
http://www.ppti.com/Technology/technology.htm
Galgali, P., Varma, A. J., Puntambekar, U. S. & Gokhale, D. V. Towards biodegradable
polyolefins: strategy of anchoring minute quantities of monosaccharides and dis-
accharides onto functionalized polystyrene, and their effect on facilitating polymer
biodegradation. Chemical Communications, 2002, 2884 - 2885, (2002).
Questions
• Explain the expression thermosetting and thermoplastic.

53
• What is the glass transition temperature?
• What is the difference between an amorphous and crystalline plastic?
• What happens to a thermoplastic when it is cooled down below melting tem-
perature?
• How do thermosetting polymers respond to subsequent heating?
• What is the difference between a saturated and an unsaturated molecule?
• Describe the two different processes of forming polymers: addition and conden-
sation.
• What are some attractive engineering properties of plastics, and in what area
do they fall?
• What are some reasons that additive agents are incorporated into plastics?
• What kind of additive agents are there, and what are their objective?
• What is the primary engineering benefit of an oriented plastic?
• What is the unique mechanical property of elastomeric materials?
• What is vulcanization?
• What is rubber?
• What is a SPI code, and what is it good for?
• What groups of SPI codes are there?
• What is a biopolymer?
• What kind of different biopolymers exist?
• What products can be made from biopolymers?
Polymers can be described as being thermosetting or thermoplastic
a. Why are the thermosetting polymers so different from thermoplastic polymers?
b. What decides if a polymer will be thermosetting or thermoplastic?
c. What are three important, common thermosetting polymers?
d. What are three important, common thermoplastic polymers?
Polymers can be described as being amorphous or crystalline polymers
a. Compare the physical and chemical properties of amorphous and crystalline
polymers.
b. What are three important amorphous polymers?
c. What are three important crystalline polymers?
d. How does light penetration qualities depend on the degree of crystallization?

54
4. Composites
Composite materials are materials that combine two or more materials (a selected
filler or reinforcing elements and compatible matrix binder) that have quite differ-
ent properties that when combined offer properties which are more desirable than
the properties of the individual materials . The different materials work together
to give the composite unique properties, but within the composite you can easily
see the different materials, they do not dissolve or blend into each other.
The key characteristic of composites is the

• Specific strength (the strength to weight ratio σ/ρ)

• Specific stiffness or specific modulus (the stiffness-to-weight ratio E/ρ)

• Tailored material ( since composites are composed of 2 or more ”phases”,

they can be formulated to meet the needs of a specific application with
considerable ease)

Composites are not a single material but a family of materials whose stiffness,
strength, density, and thermal and electrical properties can be tailored. The
matrix, the reinforcement material, the volume and shape of the reinforcement,
the location of the reinforcement, and the fabrication method etc. can all be
varied to achieve required properties.

Composite applications
Specific composite applications are detailed in each market category:
• Transportation
• Electrical/Electronics
• Building Construction
• Infrastructure
• Aerospace/Defense
• Consumer/Recreation
• Medical Products
• Sport equipment

55
4.1. The history of composites
The theory behind the construction of composite materials comes from the need
to create a strong stiff and light material.
Materials such as glass, carbon and Kevlar have extremely high tensile and
compressive strength, but in solid form, many random surface flaws present in such
materials, cause them to crack and fail at a much lower stress that it theoretically
should.
To overcome this problem, the material is produced in a fibre form, although
the flaws will occur at the same frequency, the flaws will be reduced to a small
number of fibres at any one point, and the remaining ones will carry the load with
the materials theoretical strength. To prevent flaws occurring from abrasion on
the surface of the material, or from existing flaws transferring to other fibres, it is
necessary to isolate the fibres. This is why a resin matrix system is used.
Another well-known composite is concrete. Here, aggregate (small stones or
gravel) is bound together by cement. Concrete has good strength under compres-
sion, and it can be made stronger under tension by adding metal rods, wires, mesh
or cables to the composite (so creating reinforced concrete).
Most composites are made up of just two materials. One material (the matrix
or binder) surrounds and binds together a cluster of fibres or fragments of a much
stronger material (the reinforcement). There are both natural and man-made
composites.

4.2. Natural composites

Composites exist in nature.
A piece of wood is a composite, see Figure 4.1, with long fibres of cellulose
(a very complex form of starch) held together by a much weaker substance called
lignin. Cellulose is also found in cotton and linen, but it is the binding power of
the lignin that makes a piece of timber much stronger than a bundle of cotton
fibres. Other examples of natural composites are:

Spider silk
Spider silk is a biopolymer fibre and a natural composite material, see Figure
4.2.
Its composition is a mix of an amorphous polymer (which makes the fibre
elastic), and the two simplest proteins (which give it toughness), in other words,
it is simply a protein. The result is a good combination of strength and toughness.

56
 Figure 4.1: The cell-structure of a natural composite, a tree.

It is five times as strong as steel, twice as strong as Kevlar at same weight,

twice as elastic as polyamide fibres (it can be stretched by 31% without breaking),
more elastic than aramid fibre, finer than a human hair, and lighter than cotton.
Promises are: bulletproof vests, parachute cords, bridge suspension cables,
wear-resistant shoes and clothing, seat belts, rustfree bumpers for automobiles,
artificial tendons and ligaments, etc.
Spider silk is produced either by spider or (in the future) by bacteria or plants,
therefore in an environment-friendly way. It is an environment-friendly biopolymer
and is easy to recycle. Spider silk does not decompose like other proteins by fungi,
or bacteria because they contain substances that makes it durable.
The thread of the garden spider (Araneus) would have to be 80 km long before
it would snap under its own weight. When synthesis of spider silk in bacteria or
plants becomes efficient, spider silk will have a commercial relevance.

Human hair, bone and muscles are also natural composites. These kind of
structures are also called hierarchical structures since they have structures in
many levels, see also chapter 9.4.

57
 Figure 4.2: Spidersilk.

4.3. Man-made Composites

Humans have been using composite materials for thousands of years. Take mud
bricks for example. A cake of dried mud is easy to break by bending, which puts
a tension force on one edge, but makes a good strong wall, where all the forces
are compressive. A piece of straw, on the other hand, has a lot of strength when
you try to stretch it but almost none when you crumple it up. But if you embed
pieces of straw in a block of mud and let it dry hard, the resulting mud brick
resists both squeezing and tearing and makes an excellent building material. Put
more technically, it has both good compressive strength and good tensile strength.
In the case of mud bricks, the two roles are taken by the mud and the straw; in
concrete, by the cement and the aggregate; in a piece of wood, by the cellulose
and the lignin. In fiberglass, the reinforcement is provided by fine threads or fibres
of glass, often woven into a sort of cloth, and the matrix is a plastic.
Composite materials are composed of a matrix material reinforced with any of
a variety of fibers made from ceramics, metals, or polymers. The reinforcing fibers
are the primary load carriers of the material, with the matrix component trans-
ferring the load from fiber to fiber. Reinforcement of the matrix material may be
achieved in a variety of ways. Fibers may be either continuous or discontinuous,
see Figure 4.3. Reinforcement may also be in the form of particles. The matrix
material is usually one of the many available engineering plastics/polymers. Selec-
tion of the optimal reinforcement form and material is dependent on the property
requirements of the finished part.

58
 Figure 4.3: Composite materials with different types of reinforcements.

The threads of glass in fiberglass are very strong under tension but they are
also brittle and will snap if bent sharply. The matrix not only holds the fibres
together, it also protects them from damage by sharing any stress among them.
The matrix is soft enough to be shaped with tools, and can be softened by suitable
solvents to allow repairs to be made. Any deformation of a sheet of fiberglass
necessarily stretches some of the glass fibres, and they are able to resist this, so
even a thin sheet is very strong. It is also quite light, which is an advantage in
many applications.
Over recent decades many new composites have been developed, some with
very valuable properties. By carefully choosing the reinforcement, the matrix,
and the manufacturing process that brings them together, engineers can tailor
the properties to meet specific requirements. They can, for example, make the
composite sheet very strong in one direction by aligning the fibres that way, but
weaker in another direction where strength is not so important. They can also
select properties such as resistance to heat, chemicals, and weathering by choosing
an appropriate matrix material.

59
4.3.1. Fiber-Reinforced Composites
Fiber Reinforced Plastics (FRP) is a general term for composite materials or parts
that consist of a resin matrix that contains reinforcing fibers such as glass or fiber
and have greater strength or stiffness than the resin. FRP is most often used to
denote glass fiber-reinforced plastics.
Reinforcing fibers can be made of metals, ceramics, glasses, or polymers that
have been turned into graphite and known as carbon fibers.
Fibers increase the modulus of the matrix material. The strong covalent bonds
along the fiber’s length gives them a very high modulus in this direction because
to break or extend the fiber the bonds must also be broken or moved. Fibers are
difficult to process into composites, making fiber-reinforced composites relatively
expensive.
Fiber-reinforced composites are used in some of the most advanced, and there-
fore most expensive, sports equipment, such as a time-trial racing bicycle frame
which consists of carbon fibers in a thermoset polymer matrix. Body parts of
race cars and some automobiles are composites made of glass fibers (fiberglass) or
carbon fibers in a thermoset matrix.
On the basis of stiffness and strength alone, fibre reinforced composite ma-
terials do not have a clear advantage particularly when it is noted that their
elongation to fracture is much more lower than metals with comparable strength.
The advantage of composite materials appear when the modulus per unit weight
(specific modulus) and strength per unit weight (specific strength) are considered.
The higher specific weight and high specific strength of the composite materials
means that the components weight can be reduced.
Although glass fibres are by far the most common reinforcement, many ad-
vanced composites now use fine fibres of pure carbon. Carbon fibres are much
stronger than glass fibres, but are also more expensive to produce. Carbon fibre
composites are light as well as strong. They are used in aircraft structures and in
sporting goods (such as golf clubs), and increasingly are used instead of metals to
repair or replace damaged bones. Even stronger (and more costly) than carbon
fibres are threads of boron.

4.3.2. Classification of composites

Many terms have been used to define FRP composites. Modifiers have been used
to identify a specific fiber such as Glass Fiber Reinforced Polymer (GFRP), Car-
bon Fiber Reinforced Polymer (CFRP), and Aramid Fiber Reinforced Polymer

60
(AFRP). Another familiar term used is Fiber Reinforced Plastics. In addition,
other acronyms were developed over the years and its use depended on geograph-
ical location or market use. For example, Fiber Reinforced Composites (FRC),
Glass Reinforced Plastics (GRP), and Polymer Matrix Composites (PMC) can
be found in many references. Although different, each of before mentioned terms
mean the same thing; FRP composites.
There are many ways of classifying composite materials. For instance, we can
classify the microcomposite materials based on the size, shape and distribution of
the different phases in the composite for instance like this:
• Continuous fibres in matrix: aligned, random
• Short fibres in matrix: aligned, random
• Particulates (spheres, plates, ellipsoids, irregular, hollow or solid) in matrix
• Dispersion strengthened, as for the point above, but the particle size <
10−8 m
• Lamellar structures (used in laminates)
• Skeletal or interpenetrating networks
• Multicomponent, fibres, particles etc.
Clearly, the distinction between the different groups is not always a sharp one.
Nanocomposites are composites on nanoscale (10−9 meter). They have con-
stituents that are mixed on a nanometer-length scale.
Mostly, we classify composites according to their matrix phase. The role of the
matrix is to act as a medium to keep the fibers properly oriented and to protect
them from the environment. There are ceramic matrix composites (CMC’s), metal
matrix composites (MMC’s), and polymer matrix composites (PMC’s). Materials
within these categories are often called ”advanced” if they combine the properties
of high strength and high stiffness, low weight, corrosion resistance, and in some
cases special electrical properties. This combination of properties makes advanced
composites very attractive for aircraft and aerospace structural parts.

4.4. Ceramic matrix composites CMC

Monolithic ceramics have the disadvantage of being brittle, see Figure 4.4 from:
www.ms.ornl.gov/programs/ energyeff/cfcc/iof/chap24-6.pdf.
A reinforcing phase can improve the toughness of these materials, while still
taking advantage of the matrix’s other properties such as wear resistance, hard-
ness, corrosion resistance, and temperature resistance. A wide range of ceramic

61
 Figure 4.4: Failure modes for monolithic ceramic and CFCCs.

matrix composites (CMCs) have thus been developed that combine a matrix mate-
rial with a reinforcing phase of different composition (such as alumina and silicon
carbide) or the same composition (alumina/alumina or silicon carbide/silicon car-
bide).

Figure 4.5 compares the approximate service temperature ranges of some im-
portant polymers, metals and ceramics.
The CMC market is divided into two classes; the oxide and non-oxide mate-
rials. Some of the more common oxide matrices include alumina, silica, mullite,
barium aluminosilicate, lithium aluninosilicate and calcium aluminosilicate. Ox-
ide matrices are often more mature and environmentally stable, but non-oxide
ceramics have more superior structural properties and hardness are rapidly enter-
ing the marketplace. Examples of non-oxide ceramics are silicone carbide (SiC),
silicone nitride (Si3 N4 ), boron carbide (B4 C) and aluminium nitride (AlN).
The oxide CMCs often consist of oxide fibers (for instance alumina Al2 O3 ),
while the non-oxide CMCs consist of non-oxide fibers (for instance SiC) (see also
Chapter 4.7 for more information concerning fibers). Non-oxide CMCs are more
advanced than oxide CMCs since they have higher thermal conductivity, lower

62
Figure 4.5: Service temperature ranges of some important polymers, metals and ce-
ramics.

thermal expansion and are more stable at high temperatures than oxide CMCs.
The non-oxide CMCs are used in thermally loaded components (combustor liners,
vanes, blades and heat exchangers). The oxide CMCs have good oxidation resis-
tance, alkali corrosion resistance, low dielectric constants and potentially low cost,
and are therefore used for hot gas filters, exhaust components of aircraft engines,
and in long-life, lower temperature components.
The reinforcements of CMCs can include carbides, borides and oxides. Specific
examples of reinforcements include carbon, silicon carbide, titanium diboride,
silicon nitride and alumina. Some of the more common ceramic discontinuous
reinforcement include whiskers, platelets, and particulates having compositions of
Si3 N4 , SiC, AlN, titanium dibori, boron carbide, and boron nitride.
Ceramics and CMCs can also result in fuel efficiency in heat engines because of
higher operating temperatures, and reduction or elimination of cooling systems.
The temperature requirement in such applications is not as high as in aerospace
materials applications. Other applications of CMCs include wear parts, such as
seals, nozzles, pads, liners, grinding wheels, brakes, etc. For instance, carbon fiber
reinforced carbon composites are being used in aircraft brakes.

63
4.5. Metal Matrix Composites MMC
Metal Matrix Composites (MMC’s) are increasingly found in the automotive in-
dustry. They consist of a metal such as aluminium as the matrix, and a reinforce-
ment that could be continuos fibres such as silicon carbide, graphite or alumina,
wires such as tungsten, beryllium, titanium and molybdenium, and discontinuous
materials. Metals containing ceramic particles, whiskers or (short or long) fibers
are also gaining importance. MMC are said to be materials for the demands of
the future.
When the demands for high thermal conductivity, reduced weight, heat dissi-
pation, and high strength are factors for design, MMCs represent the Next Genera-
tion of solutions for today’s electronic requirements. Metal-matrix composites are
either in use or prototyping for the Space Shuttle, commercial airliners, electronic
substrates, bicycles, automobiles, golf clubs, and a variety of other applications.
While the vast majority are aluminum matrix composites, a growing number of
applications require the matrix properties of superalloys, titanium, copper, mag-
nesium, or iron.
Regardless of the variations, however, aluminum composites offer the advan-
tage of low cost over most other MMCs. In addition, they offer excellent thermal
conductivity, high shear strength, excellent abrasion resistance, high-temperature
operation, nonflammability, minimal attack by fuels and solvents, and the ability
to be formed and treated on conventional equipment. Aluminum MMCs are ap-
plied in brake rotors, pistons, and other automotive components, as well as golf
clubs, bicycles, machinery components, electronic substrates, extruded angles and
channels, and a wide variety of other structural and electronic applications.
Compared to monolithic metals, MMCs have:
• Higher strength-to-density ratios
• Higher stiffness-to-density ratios
• Better fatigue resistance
• Better elevated temperature properties: -Higher strength
-Lower creep rate
• Lower coefficients of thermal expansion
• Better wear resistance

The advantages of MMCs over polymer matrix composites (PMCs) are:

64
 • Higher temperature capability
• Fire resistance
• Higher transverse stiffness and strength
• No moisture absorption
• Higher electrical and thermal conductivities
• Better radiation resistance
• No outgassing
• Fabricability of whisker and particulate-
reinforced MMCs with conventional metalworking equipment
Some of the disadvantages of MMCs compared to monolithic metals and poly-
mer matrix composites are:
• Higher cost of some material systems
• Relatively immature technology
• Complex fabrication methods for fiber-reinforced systems (except for casting)
• Limited service experience

4.6. Polymer Matrix Composites PMC

Polymer Matrix Composites (PMC’s) are the most common composites, and are
also known as FRP - Fibre Reinforced Polymers (or Plastics). These materials use
a polymer-based resin as the matrix, and a variety of fibres such as glass, carbon
and aramid as the reinforcement. Matrix materials are either thermosetting or
thermoplastic polymers. Reinforcing fibers are either continuous or chopped. In
general, polymer composites processing includes contracting of polymer and fibers,
shaping, controlled heating and/or reactions, and cooling. The technology of
polymer composites has been driven to a large extent by aerospace and military
applications.
Advantages of Polymer Composites
Polymer composites make the largest and most diverse use of composites due
to ease of fabrication, low cost and good properties. Also, they have:
• High specific strength properties (20-40% weight savings)
• Ability to fabricate directional mechanical properties
• Outstanding corrosion resistance
• Excellent Fatigue and fracture resistance
• Lower tooling cost alternatives
• Lower thermal expansion properties

65
 Figure 4.6:

• Simplification of manufacturing by parts integration

• Potential for rapid process cycles
• Ability to meet stringent dimensional stability requirements
Glass-fiber reinforced composites (GFRC) are strong, corrosion resistant and
lightweight, but not very stiff and cannot be used at high temperatures. Applica-
tions include auto and boat bodies, aircraft components.
Carbon-fiber reinforced composites (CFRC) use carbon fibers, which have the
highest specific module (module divided by weight). CFRC are strong, inert,
allow high temperature use. Applications include fishing rods, golf clubs, aircraft
components.

4.7. Fiber reinforcement

Fibers (see [14]) are pliable hair-like substances, built up by long chains of basic
molecules. Fibres are very small in diameter in relation to their length. Long
continuous strands of fibers are called filaments. The only natural filament fibre
is silk from butterflies or spiders. Fibre’s properties depend strongly on both
the external and internal fibre structure as well as the chemical composition.
Properties therefore vary significantly. Crystalline areas give tensile strength,
stiffness and stability, while amorphous areas are weaker but more movable.
The fibers can be individual strands as thin as a human hair or they can be
multiple fibers braided in the form of a yarn (or tow).

66
4.7.1. Choosing fibers
The fibers in advanced composites are usually non-metallic in nature, consisting
of such materials as glass, carbon, silicon carbide, alumina, boron or Kevlar.
Other fibers such as spectra and quartz are also available, but these are usually
reserved for specialized application. The table below shows some advantages and
disadvantages of some fibertypes, while Figure 4.7 shows some properties of some
fibers.
Fiber Advantages Disadvantages
E-, S-Glass - High strength - Low stiffness
- Low cost - Short fatigue life
- High temp sensitivity
Aramid (Kevlar) - High tensile strength - Low comp. strength
- Low density - High moisture absorp.
Boron - High stiffness - High cost
- High comp. strength
Carbon (AS4, T300) - High strength - Moderate cost
- High stiffness
Graphite (pitch) - Very high stiffness - Low strength
- High cost
Ceramic (silicon - High stiffness - Low strength
carbide, alumina) - High use temp - High cost

Figure 4.8 shows a plot of specific strength vs. specific stiffness of some fibers.

High-strength fibers utilized in high-performance composites include carbon,

glass, aramid, ultrahigh molecular-weight polyethylene (PE), boron, quartz, ce-
ramic and hybrid combinations. Continuous bundles (tow or roving) are the basic
fiber forms for high-performance composite applications. These forms may be used
directly in processes such as filament winding or pultrusion or may be converted
into tape, fabric and other forms.
In high-performance composites, the structural properties of the finished part
are derived primarily from the fiber. A fiber-to-matrix ratio of 60:40 or higher
is common for both thermoset- and thermoplastic-matrix composites. Desirable
properties are not achieved, however, without effective adhesion between fiber and
matrix, which requires sufficient saturation with resin (wetout) at the fiber-matrix

67
 Figure 4.7: Properties of some fibers.

Figure 4.8: Specific strength vs. specific stiffness of some fibers.

68
interface. Attention to fiber surface preparation, such as use of a surface finish
or coupling agent and selection of a compatible fiber and matrix combination,
ensures good adhesion.
We will have a closer look at some types of fibers.

4.7.2. Glass fibers

Glass fibers (GF) are the most widely used reinforcement for plastic and rubber
products, and are also the finest (smallest diameter) of all fibers, typically 1 to 4
microns in diameter. Glass fibers, see Figure 4.9, are made of silicon oxide with
addition of small amounts of other oxides. Glass fibers are characteristic for their
high strength, good temperature and corrosion resistance, and low price. Over
90% in the composite industry is glass fibers. Properties of glass fibers are bound

Figure 4.9: Glass fiber.

to the chemical composition of the mixture, but they are also influenced by the
spinning way. Usually, they are divided into:

Use Type of glass

Multipurpose fibers glass E
Acid resistant fibers glass A, C, CR
Alkali resistant fibers glass R, S
High strength fibers glass R, S
Fibers with good dielectric properties glass D

69
 The two types of glass fibers that are most common are the E-glass and S-
glass. The first type is the most used, and takes its name from its good electrical
properties (no electrical interference in thunderstorms, and no sparks). The second
type is very strong (S-glass), stiff, and temperature resistant. They are used as
reinforcing materials in many sectors, e.g. automotive and naval industries, sport
equipment. S-glass are used for instance in aircraft flooring, helicopter blades etc.
Some properties for E- and S- glass are listed in the following table:

E-glass S-glass
◦ ◦
- softens at ≈850 C - softens at ≈1000 C
- most inexpensive fiber ≈ 30% stronger than E-glass
- most commonly used fiber today ≈ 15% stiffer than E-glass
- insulators and capacitors ≈ 3 times as expensive as E-glass
- high quality glass fiber
- high technical purposes

The most widely used composite material is fiberglass in polyester resin, which
is commonly referred to as just fiberglass. Fiberglass is light weight, corrosion
resistant, economical, easily processed, has good mechanical properties, and has
over 50 years of history. It is the dominant material in industries such as boat
building and corrosion equipment, and it plays a major role in industries such as
automotive, medical, recreational, and industrial equipment.

Quartz fibers Quartz fibers (SiO2 ) are very high silica version of glass with
much higher mechanical properties and excellent resistance to high temperatures
◦
and can be used continuously to 1048.9 C and for short-temperature applications,
◦
even to 1248,9 C. They are more expensive than E-glass and S-glass, provide
significantly better electromagnetic properties than glass, making them a good
choice for parts such as aircraft radomes.

4.7.3. Carbon fibers

Carbon fibers (CF) appeared on the market in 1960 and are produced of organic
fibers (rayon, acrylics etc.) or from remaining of petroleum or tar distillation.

70
 Figure 4.10: Carbon fibre.

Carbon fibers, see Figure 4.10, are the stiffest and strongest reinforcing fibres for
polymer composites, the most used after glass fibres. Made of pure carbon in
form of graphite, they have low density and a negative coefficient of longitudi-
nal thermal expansion. Carbon fibres are very expensive and can give galvanic
corrosion in contact with metals. They are generally used together with epoxy,
where high strength and stiffness are required, i.e. race cars, automotive and space
applications, sport equipment.

4.7.4. Boron fiber

Boron (BF) is a metal known for its exceptional resistance and form. BF is five
times as strong and twice as stiff as steel. Boron provides strength, stiffness,
light weight and excellent compressive properties, as well as buckling resistance.
Use of boron composites ranges from sporting goods, such as fishing rods, golf
club shafts, skis and bicycle frames, to aerospace applications such as aircraft
empennage skins, space shuttle truss members and prefabricated aircraft repair
◦
patches. It has a melting point at 2180 C.

4.7.5. Ceramic fibers

Ceramic fibers (CEF) are mostly used as refractory fibers in uses over 1000◦ C and
are characterized by a polycrystalline structure rather than amorphous. Some
of the more common ceramic continuous reinforcements include silica, mullite,

71
alumina, carbon, zirconia, and silicon carbide in addition. The more sophisticated
kinds of fibers, have a basis of borate, carbide, silicon nitrure and borate nitrure.
They are very expensive fibers because only a small quantity is produced, and
they are used in particular fields such as aerospace.
Ceramic fibers offer high to very high temperature resistance but poor im-
pact resistance and relatively poor room-temperature properties. They are often
divided into the following two groups:
* Non-oxide fibers can be such as Siliconcarbide (SiC), boron nitride (BN),
silicon nitride (SiN), aluminium nitride (AlN) and multiphase fibers consisting of
silicon, carbon, boron, nitride, titanium, (Si-C-B-N-Ti).
* Oxide fibers can be such as aluminium oxide (Al2 O3 ), alumina zirconia mix-
tures (Al2 O3 +Zr2 O2 ), yttria-alumina-garnet (YAG), and mullite (3Al2 O3 -2SiO2 ).
Ceramic fibers of pure Al2 O3 in a single crystalline microstructure are called sap-
phire fibers.

4.7.6. Polymer Fibers

Aramid fiber (Kevlar)
Aramids are a family of nylons, including Nomex° R
and Kevlar° R
( Polyamide
-PA). Aramid fibers, see Figure 4.11 are known for their large hardness and resis-
tance to penetration. Thanks to their toughness aramid fibres are used where high

Figure 4.11: Aramid fibre.

impenetrability is required, e.g. bulletproof vests, bike tires, airplanes wings, and
sport equipment. These fibres are not as spread as glass or carbon fibres, mostly
because of their cost, high water absorption, and their difficult post-processing.

72
(References: DuPont (Kevlar° R
) Fibers of KEVLAR° R
consist of long molec-
ular chains produced from poly-paraphenylene terephthalamide). PPTA is the
acronym for para-phenylen-terephthalamid.
Polymers are not only used for the matrix, they also make a good reinforcement
material in composites. For example, Kevlar is a polymer fibre that is immensely
strong and adds toughness to a composite. It is used as the reinforcement in
composite products that require lightweight and reliable construction (e.g., struc-
tural body parts of an aircraft). Composite materials were not the original use
for Kevlar; it was developed to replace steel in radial tires and is now used in
bulletproof vests and helmets. Kevlar, and aramid-fiber composite can be used
as textile fibers. Applications include bullet-proof vests, tires, brake and clutch
◦
linings. Servicetemperature of Aramid fibre is up to 250 C, and its melting tem-
◦
perature is 426.7 C.

UHMW-PE - Ultra-high molecular weight polyethylene

UHMW, see Figure 4.12, is a polymer material composed of long chains with

Figure 4.12: UHMW.

a very high molecular weight. It even replaced Kevlar° R

for making bullet-proof
vests! UHMW-PE combines the traditional abrasion and cut resistance of metal
alloys with the impact and corrosion resistance of synthetic materials. Tough as
a metal, self-lubricating and slippery. UHMW-PE fibres have the tendency to
elongate under load (creep). They are extremely low density, and they can float.
◦ ◦
Melting-point between 120 and 150 C. Applications are being developed partic-
ularly in the areas of ballistic protection and ropes. The material is inflammable.
These fibres have a tensile strength 20 times greater than that of steel, and
40% stronger than aramid. Their free breaking length is around 330 km (steel
breaks at 25, glass at 135, carbon at 195 and aramid at 235).

73
 Figure 4.13: Liquid crystal polymer.

Products: Bullet-proof vest, Cut-resistant gloves, Psychiatric clothing.

LCP - Liquid crystal polymer

LCP, see Figure 4.13 is a thermoplastic fibre with exceptional strength and
rigidity (five times that of steel), and about 15 times the fatigue resistance of
aramid. Very good impact resistance. It doesn’t absorb moisture, has very low
stretch, it doesn’t creep like UHMW-PE fibre, and has excellent abrasion, wear,
◦
and chemical resistance. Its high melting-point (320 C) allows the retention of
these properties over broad ranges of temperatures. It is suitable for industrial,
electronic, and aerospace applications, as well as for ropes, and sport equipment.
Products: Fishing line, High performance rope, Tennis racket.

4.7.7. Carbon nanotube (CNT) fibers

Strong and versatile carbon nanotubes are finding new applications in improving
conventional polymer-based fibers and films. For example, composite fibers made
from single-walled carbon nanotubes (SWNTs) and polyacrylonitrile, a carbon
fiber precursor, are stronger, stiffer and shrink less than standard fibers.
Nanotube-reinforced composites could ultimately provide the foundation for a
new class of strong and lightweight fibers with properties such as electrical and
thermal conductivity unavailable in current textile fibers

74
 Figure 4.14: Spidersilk.

4.7.8. Fiber hybrids

Fiber hybrids capitalize on the best properties of various fiber types, while re-
ducing raw material costs. Hybrid composites combining carbon/aramid and
carbon/glass fibers have been used successfully in ribbed aircraft engine thrust
reversers, telescope mirrors, drive shafts for ground transportation and infrastruc-
ture column-wrapping systems. For example, in hybrid-fiber flywheels for electric
vehicles, the composite components withstand energy forces as high as 2.5 million
ft-lb.

4.7.9. Spider silk

Spider silk is a biopolymer fiber and a natural composite, see Figure 4.14 and
chapter 4.2. It is 5 times as strong as steel and 2 times as strong as Kevlar.
Man-made spider silk has now become a reality. "Spider-goats," ( goats that
are created having one spider gene added to the 70,000 genes that make a goat a
goat) for example, produce an extra protein in their milk that can be spun into
man-made spider silk. Spider silk is stronger, by weight, than anything else on
earth. The hope was to spin silk from milk - and use the ultra-light material for
things like bulletproof vests. It turns out that it takes a lot of goat milk to make
just a small bit of silk, so for now the idea of making bulletproof vests out of
spider silk has been put aside. Other uses of the material, medical sutures, for
example are in progress instead.

75
 Figure 4.15: Crayons on Aerogel over a flame.

4.8. Aerogel composites

Aerogel is 99.8% air, and provides 39 times more insulating than the best fiberglass
insulation, and is 1,000 times less dense than glass, see Figure 4.15 http://stardust.jpl.nasa.gov/
tech/aerogel.html where the picturecan be found. A cube of 1x1x1 meter of aero-
gel will only weight three kg. It can also be exposed to temperatures up to 1400◦ C
before it will soften.
Aerogel is not like conventional foams, but is a special porous material with
extreme microporosity on a micron scale. It is composed of individual features
only a few nanometers in size. These are linked in a highly porous dendritic-like
structure.
This exotic substance has many unusual properties, such as low thermal con-
ductivity, refractive index and sound speed - in addition to its exceptional ability
to capture fast moving dust. Aerogel is made by high temperature and pressure-
critical-point drying of a gel composed of colloidal silica structural units filled with
solvents.
Silica aerogel is just another form of glass. If aerogel is handled roughly, it will
break just like glass. However, if care is taken, the material can be handled and
shaped effectively.
Silica aerogels are inorganic solids, very brittle and usually very low density
materials. A collapse of the solid network in the material occurs gradually, spread-
ing the force of impact out over a longer time, and is why they can be very useful
as energy absorbing materials.
Aerogels can be added different materials/particles in order to obtain special

76
effect of the materials. For instance can we add metal salts, or other compounds
to a sol before gelation. Then we may obtain a deep blue aerogel by adding nickel;
while a pale green color appears by adding copper; a black gel contains carbon
and iron; and orange aerogel is added iron oxide. It is also possible to obtain a
magnetic aerogel by using chemical vapor infiltration to get iron oxide introduced
in the material. The aerogel can also be made photoluminescent, which means
that it glows-in-the-dark by using a pigment that absorbs light and emits that
energy over a period of time -even after the light source has been removed. The
pigment is rechargeable.

Flexible aerogels
In the mid to late 1990’s, Aspen Systems perfected proprietary changes in
the formulation, processing and drying of aerogels, reducing drying time to a few
hours. Besides a cost effective drying time, the aerogels were also isolated in the
form of thin, flexible blankets. The blankets were much more robust than the
monoliths and beads, and could also be easily installed like any other flexible bat-
ting. These breakthroughs led to the formation of Aspen Aerogels, Inc. in 2001,
and the commercialization of aerogel technology for broad use, see for instance
the Figure 4.16
Aspen Aerogels has created a variety of organic and inorganic/organic hy-
brid aerogel formulations including those made from polydimethylsiloxane/silica,
cellulose polyurethane, polyimide, polymethylmethacrylate/silica, and polybuta-
diene rubber. These materials show enhanced physical and mechanical properties
relative to pure silica aerogel.
As shown in the table below, aerogels offer significantly more insulating value
per unit of material than other insulations:
W
Material Thermal Conductivity (k) [ mK ]
Aerogel 0.012
Polyurethane foam 0.021
Polystyrene foam 0.038
Microporous silica 0.019-0.038
fiberglass 0.040
Mineral wool 0.038
Perlite 0.040-0.060
Calcium Silicate 0.047

77
Figure 4.16: Areas that Aspens flexible aerogels can be used in. The picture is from:
http://www.aerogel.com/markets.htm

4.9. Textile composites

There is a new ”class” of composites, called textile composites. The material
comes from a new technique which instead of the reinforcing fibres being put in
place individually, which is slow and costly, they can be knitted or woven together
to make a sort of cloth. This can even be three-dimensional rather than flat, see
Figure 4.17. The spaces between and around the textile fibers may be filled with
the matrix material (such as a resin) to make the final product.
This process can quite easily be done by machines rather than by hand, making
it faster and cheaper. Connecting all the fibres together also means that the
composite is less likely to be damaged when struck.

4.10. Bio-inspired materials

Researchers are making rapid progress in the design and synthesis of non-natural
oligomers and polymers that emulate the properties of natural proteins (also called
bio-inspired materials). For instance is NASA trying to tailor new materials that
can mimic the extraordinary structural and self-repairing properties of biological

78
 Figure 4.17: Three dimensional braiding of composite structures.

substances such as bone or sea shells, also called bio-inspired materials. Such
biologically inspired materials can adapt to changing conditions and therefore
they can help to make airplanes and spacecraft lighter, stronger and more reliable.
New advanced materials can also be able to change their shapes (i.e. airplane
wings which today is done with hydraulics). The hope is to find materials that
change shape on command because then we could improve for instance airplanes
tremendously.
Also the so-called ’self-healing’ materials could be very important to space
exploration, thus even small sand grain particle of a meteor could puncture the
hull of existing space vehicles.

4.11. Self-healing composites

Inspired by biological systems in which damage triggers an autonomic healing re-
sponse, researchers at the University of Illinois have developed a synthetic material
that can heal itself when cracked or broken.
The material — consisting of a microencapsulated healing agent and a special
catalyst embedded in a structural composite matrix — could increase the reliability
and service life of thermosetting polymers used in a wide variety of applications
ranging from microelectronics to aerospace.
”Once cracks have formed within typical polymeric materials, the integrity of
the structure is significantly compromised,” said Scott White, a UI professor of
aeronautical and astronautical engineering and lead author of a paper published in
the Feb. 15 (2001) issue of the journal Nature that described the new self-healing
material.

79
 Figure 4.18: Self healing composite from http://www.cerncourier.com.

”Often these cracks occur deep within the structure where detection is diffi-
cult and repair is virtually impossible.” In the new material, however, the repair
process begins as soon as a crack forms.
”When the material cracks, the microcapsules rupture and release the healing
agent into the damaged region through capillary action,” White said. ”As the
healing agent contacts the embedded catalyst, polymerization is initiated which
then bonds the crack face closed.”
In recent fracture tests, the self-healed composites recovered as much as 75
percent of their original strength. And because microcracks are the precursors to
structural failure, the ability to heal them will enable structures that last longer
and require less maintenance.
The concept of the autonomic healing system.
”Filling the microcracks will also mitigate the harmful effects of environ-
mentally assisted degradation such as moisture swelling and corrosion cracking,”
White said. ”This technology could increase the lifetime of structural components,
perhaps by as much as two or three times.”
The ability to self-repair and restore structural integrity also could extend the
lifetimes of polymer composite circuit boards, where microcracks can lead to both
mechanical and electrical failure.

80
Figure 4.19: (a) In a right-handed material, where both permittivity and permeability
are positive, the index of refraction is positive and the law of optics follow our intuition.
(b) This is not the case in a left-handed material with negative permittivity and perme-
ability. In that case the index of refraction is negative and a convergent lens becomes
divergent. Figure from: http://www.ifh.ee.ethz.ch/~martin/res10.en.html

4.12. Left-handed metamaterials

Metamaterials are engineered composites exhibiting properties that are not ob-
served in the constituent materials and not observed in nature.
The response of a material to electromagnetic fields is entirely characterized by
its permittivity and its permeability. The former gives the response to an applied
electric field, while the latter describes the response to a magnetic field. They
determine how electromagnetic waves (including microwaves and visible light)
interact with the material. These two parameters are not constant, but depend
on the illumination frequency (for example a semiconductor can be opaque in
the visible range, but transparent in the infrared). Quite surprising effects can
also arise for specific values of these two parameters. For example, when the
permittivity takes particular negative values plasmon resonances can be excited
in the material, see Figure 4.19.

Composite Material with ”reversed” physical properties

Minneapolis, MN-Physicists at the University of California, San Diego (UCSD)
have produced a new class of composite materials with unusual physical properties
that scientists theorized might be possible, but have never before been able to
produce in nature.
”Composite materials like this are built on a totally new concept,” said the
two co-leaders of the UCSD team, Sheldon Schultz and David R. Smith, who

81
announced their discovery at a news conference. ”While they obey the laws of
physics, they are predicted to behave totally different from normal materials and
should find interesting applications.”
The unusual property of this new class of materials is essentially its ability
to reverse many of the physical properties that govern the behavior of ordinary
materials. One such property is the Doppler effect, which makes a train whistle
sound higher in pitch as it approaches and lower in pitch as it recedes. According
to Maxwell’s equations, which describe the relationship between magnetic and
electric fields, microwave radiation or light would show the opposite effect in this
new class of materials, shifting to lower frequencies as a source approaches and to
higher frequencies as it recedes.
Similarly, Maxwell’s equations further suggest that lenses that would normally
disperse electromagnetic radiation would instead focus it within this composite
material. This is because Snell’s law, which describes the angle of refraction
caused by the change in velocity of light and other waves through lenses, water
and other types of ordinary material, is expected to be exactly opposite within
this composite.
”If these effects turn out to be possible at optical frequencies, this material
would have the crazy property that a flashlight shining on a slab can focus the
light at a point on the other side,” said Schultz. ”There’s no way you can do that
with just a sheet of ordinary material.”
Underlying the reversal of the Doppler effect, Snell’s law, and Cerenkov ra-
diation (radiation by charged particles moving through a medium) is that this
new material exhibits a reversal of one of the ”right-hand rules” of physics which
describe a relationship between the electric and magnetic fields and the direction
of their wave velocity.
The new materials are known by the UCSD team colloquially as ”left-handed
materials,” after a term coined by Veselago, because they reverse this relationship.
What that means is physically counterintuitive-pulses of electromagnetic radiation
moving through the material in one direction are composed of constituent waves
moving in the opposite direction.
The UCSD physicists emphasized that while they believe their new class of
composites will be shown to reverse Snell’s law, the specific composite they pro-
duced will not do so at visible-light frequencies. Instead, it is now limited to
transmitting microwave radiation at frequencies of 4 to 7 Gigahertz-a range some-
where between the operation of household microwave ovens (3.3 Gigahertz) and
military radars (10 Gigahertz).

82
 The composite constructed by the UCSD team-which also consisted of Willie
J. Padilla, David C. Vier, and Syrus C. Nemat-Nasser-was produced from a series
of thin copper rings and ordinary copper wire strung parallel to the rings. It is
an example of a new class of materials scientists call ”metamaterials.” ”Even
though it is composed of only copper wires and copper rings, the arrangement has
an effective magnetic response to microwaves that has never been demonstrated
before,” said Schultz.
What’s unusual about the new class of materials produced by the UCSD team
is that it simultaneously has a negative electric permittivity and a negative mag-
netic permeability, a combination of properties never before seen in a natural or
man-made material.
”And the interesting thing is that it’s produced with no magnetic material,”
said Schultz. ”It’s all done with copper.”
”The bottom line,” said Smith, ”is that this material- this metamaterial, at
frequencies where both the permittivity and permeability are negative, behaves
according to a left-handed rule, rather than a right-handed rule.”

Questions
• What is the theory behind composite materials?
• What is a composite material?
• How do we classify composite materials according to their matrix phase?
• What is the greatest advantage of composite materials?
• What is the greatest disadvantage of composite materials?
• Why is specific modulus and specific strength important?
• What does Fiber Reinforced Plastics (FRP) mean?
• What kind of reinforcement can composites have? Illustrated them.
• What kind of fibers are there, and what are their advantages/disadvantages?
• What is so special about spider silk, can you find out the prospect for commer-
cialisation of spider silk?
• What is the primary function of the reinforcement and the matrix in the com-
posite material?
• How do we find the modulus of the entire composite?
• How do we classify composites?
•What is so special about high performance materials?
• What is the strength of the composite primarily depending on?
• What is CMC, MMC, and PMC? What are the differences? Where are the
applications?

83
• What is the advantages of MMCs compared to monolithic metals?
• What is the advantages of MMCs compared to PMC’s?
• What is the disadvantages of MMCs compared to PMC’s and monolithic metals?
• How does a self-healing composite work?
• What is a left-handed material?
• What is a metamaterial?

References:
Natural composites:
http://www.imb-jena.de/~sponner/Current.html
Fibers
http://www.azonano.com/news.asp?newsID=82
MMC:
http://www.tms.org/pubs/journals/JOM/0104/Rawal-0104.html
http://www.electrovac.com/metall/indexe.htm
http://www.machinedesign.com/bde/materials/compsites/rvmat2d.html
http://www.pcc-aft.com/tech/mmc/mmc.html
Ceramics:
http://www-materials.eng.cam.ac.uk/mpsite/properties/non-IE/max_service_temp.html
http://www.albint.com/web/techweav/techw.nsf/
http://www.engr.utk.edu/~cmc/
http://www.onera.fr/dmsc-en/matcer/
http://www.mmat.ubc.ca/other/courses/mmat382/cnc63.htm
http://books.nap.edu/books/0309059968/html/7.html#pagetop
http://www.ms.ornl.gov/programs/energyeff/cfcc/iof/chap24-6.pdf
PMC
http://www.npl.co.uk/npl/cmmt/cog/cmmthm098.html
http://www.egr.msu.edu/cmsc/nsf/polymeric.html
http://www.raypubs.com
http://www.raypubs.com/hpcsb/sbover2.html#partdesign
High-performance composites:
http://www.etcusa.com/hpc/
Material database on internet:
http://www.matweb.com/index.asp?ckck=1

84
Composites:
http://composite.about.com/mbody.htm
http://composite.about.com/cs/aboutcomposites/
http://www.mil17.org/
http://www.netcomposites.com/
http://www.science.org.au/nova/059/059key.htm
http://www.efunda.com/materials/polymers/properties/
polymer_cat.cfm?MajorID=PA
http://www.sdplastics.com/
http://www.howstuffworks.com/ question87.htm
http://www.users.globalnet.co.uk/~weeks/Composite%20Materials.htm
http://callisto.my.mtu.edu/my472/
http://home.earthlink.net/~ttc/about.html
http://www.mech.utah.edu/~rusmeeha/labNotes/composites.html
http://ice.chem.wisc.edu/materials/composite.html
http://www.psrc.usm.edu/macrog/index.htm
http://callisto.my.mtu.edu/MY472/
http://www.matweb.com/searchcompenglish.htm
http://plastics.about.com/library/weekly/aa980323.htm
http://www.advancedcomposites.com/technology.htm
http://www.nap.edu/books/0309059968/html/index.html
http://www.technica.net/NF/NF2/efibreinorganiche.htm
http://www.fibreglast.com/
Aerogel materials:
http://www.ntnu.no/gemini/2002-05/16-19.htm
http://www.tu.no/snadder/article.jhtml?articleID=14148
http://stardust.jpl.nasa.gov/tech/aerogel.html
http://www.taasi.com/what.htm
http://www.aspensystems.com/
http://www.aerogel.com/technology.htm
Textile composites:
http://www.mtm.kuleuven.ac.be/Research/C2/poly/Modelling.htm
http://www.muratec.net/braider/
http://www.engr.ukans.edu/~rhale/textiles/index.htm
http://www.ntcresearch.org/pdf-rpts/AnRp00/I00-A06.pdf
http://www.innovationave.com/storage/PDF/advmaterials/ghosh.pdf

85
http://fiberarchitects.com/
http://www.mtm.kuleuven.ac.be/Research/C2/poly/wisetex.html
http://www.stru.polimi.it/Compositi/Compositi_dx.htm
John Summerscales, Microstructural characterisation of fibre-reinforced compos-
ites, ISBN 1 85573 240 8, 320 pages, 1998.
General:
http://claymore.engineer.gvsu.edu/~jackh/eod/
Race-car:
http://www.mse.cornell.edu/courses/engri111/racecar.htm
FRP manufact:
http://www.owenscorning.com/owens/composites/about/introduction.asp
Sspace-age materials:
http://www.spacedaily.com/news/materials-02zh.html
High-performance composites:
http://www.etcusa.com/hpc/
http://www.raypubs.com/hpcsb/sboverview.html
http://www.atlcomposites.com/products_composite_duflex_intro.htm
http://www.quadrantepp-europe.com/idqua001.asp
http://www.bayplastics.co.uk/
http://www.tstar.com/pdf/quadrant-DF.pdf
High-performance materials:
http://www.hiper-group.com/hcproducts.htm
Self-healing material:
http://www.news.uiuc.edu/scitips/01/0214selfheal.html
http://www.globaltechnoscan.com/21stFeb-27thFeb01/biological.htm
Left-handed composite materials/metamaterials:
http://ucsdnews.ucsd.edu/newsrel/science/mccomposite.htm
http://www-physics.ucsd.edu/lhmedia/
http://www.ing.dk/konf/root/redproduktion/sub/graenseland/html/0106.html
http://physics.ucsd.edu/~drs/left_home.htm
http://www.ifh.ee.ethz.ch/~martin/res10.en.html
http://physics.ucsd.edu/lhmedia/

86
5. Smart (intelligent) Materials and Structures
Smart materials are not smarter than you. Smart materials respond to environ-
mental stimuli with particular changes in some variables. For that reason they
are often also called responsive materials or functional materials. Depending on
changes in some external conditions, ”smart” materials change either their prop-
erties (mechanical, electrical, appearance), their structure or composition, or their
functions. They are materials able to transform other forms of energy to mechan-
ical energy and, sometimes, vice versa. These are unique type of materials and
have similarities in microstructures and deformation mechanisms. Fundamental
understanding on the behavior of these materials is still on the way, meanwhile,
various novel applications of these materials are found of successful.
The technological field of smart materials (see [16]) has evolved over the past
decades with increasing pace during the 1990s.
At a more sophisticated level, such smart materials become intelligent when
they have the ability to respond intelligently and autonomously to dynamically-
changing environmental conditions.
Potential applications are widespread and have excited interest in industrial,
military, commercial, medical, automotive and aerospace fields. Embedded fibre-
optic sensing systems are employed in many engineering disciplines to monitor
critical characteristics. Several smart skins programmes have been initiated for
both civil and military aircraft. Large space structures are also candidates for the
incorporation of smart structural systems because of the variable service condi-
tions in which they operate. Typical applications include, sensors and actuators,
vibration suppression and damping device, micro-electro-mechanical-system, bio-
medical engineering, space, robot, etc., to name a few. These materials will greatly
help us to further improve the way of living of our human being (see i.e.[19]).
Furthermore, one should note that the terms Smart materials, Intelligent Ma-
terials, Active Materials, Adaptive Materials, (and to some extent) actuators and
sensors, are almost always, used interchangeably. We will consider the following
4 main groups of smart materials in more detail
• — Color Changing Materials
— Light Emitting Materials
— Moving Materials
— Temperature Changing Materials

87
5.1. Color Changing Materials
Color changing materials are materials that change color due to different external
stimuli.

5.1.1. Photochromic materials

Photochromic materials change reversibly color with changes in light intensity,
see Figure 5.1. Usually, they are colorless in a dark place, and when sunlight or
ultraviolet radiation is applied molecular structure of the material changes and it
exhibits color. When the relevant light source is removed the color disappears.
Changes from one color to another color are possible mixing photochromic
colors with base colors. They are used in paints, inks, and mixed to mould or

Figure 5.1: Photochromic material.

casting materials for different applications.

5.1.2. Thermochromic materials

Thermochromic materials change reversibly color with changes in temperature,
see Figure 5.2. They usually are semi-conductor compounds. The change in color
happens at a determined temperature, which can be varied doping the material.
They are used to make paints, inks or are mixed to moulding or casting ma-
terials for different applications.

5.2. Light Emitting Materials

Light emitting materials are materials that change the light in some sense, due to
some external stimuli.

88
 Figure 5.2: Thermochromic material.

5.2.1. Electroluminescent materials

Electroluminescent materials produce a brilliant light of different colors when
stimulated electronically, see Figure 5.3.

Figure 5.3: Electroluminescent material.

Electroluminescent materials are a combination of phosphorous and fluorocar-

bons that produce a brilliant light of different colors when stimulated electronically
(e.g. by AC current). While emitting light no heat is produced.
They can be used for making lightstripes for decorating buildings, or for in-
dustrial and public vehicles safety precautions.

5.2.2. Fluorescent materials

Fluorescent materials produce visible or invisible light as a result of incident light
of a shorter wavelength (i.e. X-rays, UV-rays, etc.). The effect ceases as soon as

89
the source of excitement is removed, see Figure 5.4.

Figure 5.4: Fluorescent material.

Fluorescent pigments in daylight have a white or light color, whereas under

excitation by UV radiation they irradiate an intensive fluorescent color.
They can be used for paints, inks or mixed to moulding or casting materials
for different applications.

5.2.3. Phosphorescent materials

Phosphorescent materials (or afterglow materials) produce visible or invisible light
as a result of incident light of a shorter wavelength (i.e. X-rays, UV-rays, etc.),
detectable only after the source of the excitement has been removed, see Figure
5.5.

Figure 5.5: Phosphorescent material.

90
 Afterglow effect pigments are polycrystalline inorganic zinc sulphide (green
afterglow) or alkaline earth sulphides (red or blue afterglow), and can be used in
paints, inks or mixed to moulding or casting materials for different applications.

5.3. Moving Materials

Moving materials are materials that in some sense move when exposed to some
special external source.

5.3.1. Conducting polymers

Conducting polymers are conjugated polymers, through which electrons can move
from one end of the polymer to the other. The most common are polyaniline

Figure 5.6: Conducting polymer.

(PAni) and polypyrrole (PPY). Polipyrrole has been used for the development
of micro muscles. Polyaniline films sandwiched around a ion-conducting film are
considered as material for artificial muscles for robots. A current flow reduces one
side and oxidizes the other. Ions are transferred. One side expands and the other
contracts, resulting in a bending of the sandwich, see Figure 5.6. Electrical and
chemical energies are in this way transformed into mechanical energy.
Conducting polymers are still at a research level. Presently, the expected
lifetime of the muscle is of 100.000 actuations. In the Year 2000 Alan Heeger got
the Nobel Price in chemistry for the invention of a conducting polymer.

91
5.3.2. Dielectric elastomers
Dielectric elastomers (also called electrostrictive polymers) exhibit a mechanical
strain when subjected to an electric field as shown in Figure 5.7. The most com-

Figure 5.7: Dielectric elastomers.

mon are PMMA-based electrostrictive polymers. Thanks to their electrostrictive

strain, they can be sandwiched between two electrodes to emulate the operation of
muscles. In an electric field, the elastomer expands in the plane of the electrodes,
amplifying the normal compression due to the electrostatic charges on the elec-
trodes. The result is a muscle with large strain capability, and a large actuation
pressure.

5.3.3. Piezoelectric materials

Piezoelectric materials produce an electric field when exposed to a change in di-
mension caused by an imposed mechanical force (piezoelectric or generator effect),
see Figure 5.8. Conversely, an applied electric field will produce a mechanical
stress (electrostrictive or motor effect). They transform energy from mechanical
to electrical and vice-versa and are used for sensing purposes (e.g. microphone,
transducer), and for actuating applications. Similar to piezoelectric materials are
electrostrictive and magnetostrictive materials used in high precision actuation.
They are ferromagnetic materials which experience an elastic strain when sub-
jected to an electric or magnetic field respectively.
Products: Smart skis, Mothra plain, Acoustic transducer, Multilayer cofired
actuator, Magnetostrictive linear motor, Smart skin.

92
 Figure 5.8: Piezoelectric material.

5.3.4. Polymer gels

Polymer gels consist of a cross-linked polymer network inflated with a solvent
such as water, see Figure 5.9. They have the ability to reversibly swell or shrink

Figure 5.9: Polymer gels.

(up to 1000 times in volume) due to small changes in their environment (pH,
temperature, electric field). Micro sized gel fibres contract in milliseconds, while
thick polymers layers require minutes to react (up to 2 hours or even days). They
have high strength and can deliver sizeable stress (approximately equal to that of
human muscles).
The most common are polyvinylalcohol (PVA), polyacrylicacid (PAA) and

93
polyacrylonitrile (PAN). There are many potential applications for these mate-
rials (e.g. artificial muscles, robot actuators, adsorbers of toxic chemicals), but
presently, few of them have a commercial diffusion. Response time is still not
fast enough for artificial muscles. Lifetime of a gel actuator is very short. Their
structure gradually degrades and they become unusable. Commercial reality is
still far away.

5.3.5. Shape memory materials (SMM)

Shape memory alloys (SMA) are metals that, after being strained, at a certain
temperature revert back to their original shape, see Figure 5.10. A change in their
crystal structure above their transformation temperature causes them to return to
their original shape. SMAs enable large forces (generated when encountering any
resistance during their transformation) and large movements actuation, as they
can recover large strains.

Figure 5.10: Shape memory alloys (SMA).

Shape-memory materials are also superelastic, namely they are able to sustain
a large deformation at a constant temperature, and when the deforming force
is released they return to their original undeformed shape. Typically, they can
undergoe elastic strain up to 10%.
The shape memory effect is a unique property of certain alloys. Even though
the alloy is deformed in the low temperature phase, it recovers its original shape on
being heated to a critical higher temperature. The material is one that undergoes
a change of crystal structure at a certain temperature called the transformation
temperature. Above this temperature the material has one crystal structure (cubic

94
in the case of Cu-Al-Ni) and below this temperature it has another (orthorhombic
for Cu-Al-Ni). The low temperature structure of these types of materials allows
the material to be easily and apparently permanently deformed. However on
heating the material returns to its high temperature structure which has only
one possible shape. Thus the material has ”remembered” its shape (see also
[20]). Generally, these materials can be plastically deformed at some relatively
low temperature, and upon exposure to some higher temperature will return to
their shape prior to the deformation. Materials that exhibit shape memory only
upon heating are referred to as having a one-way shape memory. Some materials
also undergo a change in shape upon recooling. These materials have a two-way
shape memory.
Products: SMA stent for veins, Superelastic glasses, Coffeepot thermostat,
Electrical connector.
Also shape-memory ceramics and polymers (SMP) exist. Shape-memory poly-
mer for instance, converts from a temporary shape (top) to its parent shape (bot-
tom) in 45 seconds at 65o C.
Commercial SME Alloys
The only two alloy systems that have achieved any level of commercial ex-
ploitation are the NiTi alloys and the copper-base alloys. Properties of the two
systems are quite different. The NiTi alloys have greater shape memory strain (up
to 8% versus 4 to 5% for the copper-base alloys), tend to be much more thermally
stable, have excellent corrosion resistance compared to the copper-base alloys’
medium corrosion resistance and susceptibility to stress-corrosion cracking, and
have much higher ductility. On the other hand, the copper-base alloys are much
less expensive, can be melted and extruded in air with ease, and have a wider
range of potential transformation temperatures. The two alloy systems thus have
advantages and disadvantages that must be considered in a particular application.

5.3.6. Nanostructured Shape Memory Materials

Nanoscale materials have enjoyed recent interest because of the changes that oc-
cur in material behavior as surface effects and grain boundary behavior play a
larger role in the overall mechanical behavior of the material. Shape memory
alloys continue to be explored for more dramatic superelastic behavior and sev-
eral of these alloys have already proven to have wide ranging applications in the
aerospace, biomedical, and microelectronics industries. A nanostructured shape
memory alloy is on the leading edge of materials development and will allow fur-

95
Figure 5.11: Magnetically controlled shape memory materials (MSM) replace machines.

ther advances in the application of shape memory alloys to microscopic structures

and miniaturized devices.
Foils of nickel-titanium alloy with nanoscale structure have been successfully
fabricated. Preliminary characterization testing has been conducted which con-
firms that the primary phase present in the material is NiTi which is the phase
displaying shape memory behavior.

5.3.7. Magnetic Shape Memory (MSM) Materials

Magnetic Shape Memory effect is a new invention in actuator materials field, al-
lowing even 50 times greater strains than in previous magnetically controlled ma-
terials (magnetostrictive materials). In MSM materials the magnetic field moves
microscopical parts of the material (so called twins) that creates a net shape
change of the material. The mechanism enables also more complicated shape
changes than conventional linear strain, such as bending and shear. Strains can
be even 5 % (depending on the material). Typically, present MSM materials pro-
duce 2 % strain at 0 to 2 MPa stress in actuator use. AdaptaMat is developing a
commercial grade MSM alloy. MSM material samples can be produce for research
use.
MSM materials field of application seems to be very wide, ranging from auto-
motive applications to home electronics. They can simplify complicated mechan-
ical structures. Consider a sewing machine, as in Figure 5.11: traditional sewing
machine has a rotating electric motor and a large number of parts in a rather
complicated mechanical transmission system, although the desired motion of the
needle is up and down. A sewing machine based on MSM technology could consist

96
of an electromagnet (coil) and a needle made of MSM-material.

5.4. Temperature Changing Materials

Temperature changing materials are materials that change temperature in some
way, when exposed to some sort of external source.

5.4.1. Thermoelectric materials

Thermoelectric materials are special types of semiconductors that, when coupled,
function as a ”heat pump”, see Figure 5.12. By applying a low voltage DC power

Figure 5.12: Thermoelectric materials.

source, heat is moved in the direction of the current (+ to -). Usually, they are
used for thermoelectric modules where a single couple or many couples (to obtain
larger cooling capacity) are combined. One face of the module cools down while
the other heats up, and the effect is reversible. Thermoelectric cooling allows for
small size and light devices, high reliability and precise temperature control, and
quiet operation. Disadvantages include high prices and high operating costs, due
to low energy efficiency.

97
 Figure 6.1: Aluminium foam / functional gradient material in car.

6. Functional Gradient Materials

Functional (FGM) gradient materials are materials that have a gradual variation
of material properties from one end to another, as shown in Figure 6.1.
The concept of Functionally Gradient Materials (FGM) was first put for-
ward for reducing thermal stress. A kind of special materials which can be used
long-term in high temperature and large difference of temperature is needed as
aerospace industry is developing at high speed. The thermal stress in materials is
so large that neither pure metals, pure ceramics nor cladding materials can meet
this strict condition.
Man-made functionally graded materials (FGMs) copy natural biomaterials,
such as bamboo, teeth and shell etc., with a gradient in chemical composition or
microstructure from one side to the other side in the material. Materials with a
gradual change in chemical composition or microstructure are expected to have
advanced or new properties. The concept of functionally graded materials was
proposed in 1984 by material scientists in the Sendai area in Japan. Since then
studies to develop high-performance heat-resistant materials using functionally
graded technology have continued. In recent years, another attempt to apply the
FGM concept to the functional materials has been initiated.
The typical design of the FGMs is that the varying composition of the FGMs
from high ceramic content on one side of bulk material to high metal content on
the other side. This may lead to good wear resistance, good corrosion resistance,
good heat resistance on the high ceramic content side and good fracture resistance,

98
 Figure 6.2: Functional (FGM) gradient materials.

good thermal conductivity on the high metal content side.

The joining of ceramics to metals is of great engineering significance for vari-
ous advanced technologies, including high-temperature structural applications in,
e.g., turbine airfoils and combustors. A crucial matter in the ceramic-metal joint
system is to develop an efficient means to suppress the damage to be induced by
thermal stresses (cracking, interfacial decohesion, etc.), see Figure 6.2.

99
Figure 6.3: Functional (FGM) gradient materials.

100
Figure 7.1: The picture shows a solar cell which is made from a monocrystalline silicon
wafer.

7. Solar Cell Materials

A solar cell or photovoltaic cell is a device that converts light energy into electrical
energy. Sometimes the term solar cell is reserved for devices intended specifically
to capture energy from sunlight, while the term photovoltaic cell is used when the
light source is unspecified.
Fundamentally, the device needs to fulfill only two functions: photogeneration
of charge carriers (electrons and holes) in a light-absorbing material, and separa-
tion of the charge carriers to a conductive contact that will transmit the electricity
(simply put, carrying electrons off through a metal contact into a wire or other
circuit). This conversion is called the photovoltaic effect, and the field of research
related to solar cells is known as photovoltaics.
Solar cells have many applications. They have long been used in situations
where electrical power from the grid is unavailable, such as in remote area power
systems, Earth-orbiting satellites and space probes, consumer systems, e.g. hand-
held calculators or wrist watches, remote radiotelephones and water pumping
applications. More recently, they are starting to be used in assemblies of so-
lar modules (photovoltaic arrays) connected to the electricity grid through an
inverter, often in combination with a net metering arrangement (ref. wikipedia).

7.1. Light-absorbing materials

All solar cells require a light absorbing material contained within the cell struc-
ture to absorb photons and generate electrons via the photovoltaic effect. The
materials used in solar cells tend to have the property of preferentially absorbing
the wavelengths of solar light that reach the earth surface; however, some solar
cells are optimized for light absorption beyond Earth’s atmosphere as well. Light
absorbing materials can often be used in multiple physical configurations to take

101
advantage of different light absorption and charge separation mechanisms. Many
currently available solar cells are configured as bulk materials that are subse-
quently cut into wafers and treated in a "top-down" method of synthesis (silicon
being the most prevalent bulk material). Other materials are configured as thin-
films (inorganic layers, organic dyes, and organic polymers) that are deposited
on supporting substrates, while a third group are configured as nanocrystals and
used as quantum dots (electron-confined nanoparticles) embedded in a supporting
matrix in a "bottom-up" approach. Silicon remains the only material that is well-
researched in both bulk and thin-film configurations. The following is a current
list of light absorbing materials, listed by configuration and substance-name:

7.2. Silicon
By far, the most prevalent bulk material for solar cells is crystalline silicon (ab-
breviated as a group as c-Si), also known as "solar grade silicon". Bulk silicon
is separated into multiple categories according to crystallinity and crystal size in
the resulting ingot, ribbon, or wafer.

• monocrystalline silicon (c-Si): often made using the Czochralski process.

Single-crystal wafer cells tend to be expensive, and because they are cut
from cylindrical ingots, do not completely cover a square solar cell module
without a substantial waste of refined silicon. Hence most c-Si panels have
uncovered gaps at the corners of four cells.

• Poly- or multicrystalline silicon (poly-Si or mc-Si): made from cast square

ingots – large blocks of molten silicon carefully cooled and solidified. These
cells are less expensive to produce than single crystal cells but are less effi-
cient.

• Ribbon silicon: formed by drawing flat thin films from molten silicon and
having a multicrystalline structure. These cells have lower efficiencies than
poly-Si, but save on production costs due to a great reduction in silicon
waste, as this approach does not require sawing from ingots.

• New Structures: These new compounds are special arrangements of silicon

that can dramatically improve efficiency such as ormosil.

[edit]

102
7.3. Thin films
The various thin-film technologies currently being developed reduce the amount
(or mass) of light absorbing material required in creating a solar cell. This can
lead to reduced processing costs from that of bulk materials (in the case of silicon
thin films) but also tends to reduce energy conversion efficiency, although many
multi-layer thin films have efficiencies above those of bulk silicon wafers.
• CdTe: Cadmium telluride is an efficient light-absorbing material for thin-
film solar cells. CdTe is easier to deposit and more suitable for large-scale
production. Despite much discussion of the toxicity of CdTe-based solar
cells, this is the only technology (apart from amorphous silicon) that can be
delivered on a large scale.
• CIGS: are multi-layered thin-film composites. The abbreviation stands for
copper indium gallium diselenide. The best efficiency of a thin-film solar
cell as of December 2005 was 19.5% with CIGS absorber layer. Higher
efficiencies (around 30%) can be obtained by using optics to concentrate the
incident light.
• CIS: CIS is an abbreviation for general chalcopyrite films of copper indium
selenide (CuInSe2), CIGS mentioned above is a variation of CIS. While these
films can achieve 13.5% efficiency, their manufacturing costs at present are
high when compared with a silicon solar cell but continuing work is leading
to more cost-effective production processes
• Gallium arsenide (GaAs) multijunction:High-efficiency cells have been devel-
oped for special applications such as satellites and space exploration. These
multijunction cells consist of multiple thin films produced using molecular
beam epitaxy. A triple-junction cell, for example, may consist of the semi-
conductors: GaAs, Ge, and GaInP2. GaAs multijunction devices are the
most efficient solar cells to date, reaching a record high of 40.7% efficiency
under solar concentration and laboratory conditions. In production, GaAs
triple-junction cells reach efficiencies above 28.3%. They are also some of
the most expensive cells per unit area (up to US$40/cm2 ).
• Dye-sensitized solar cells are a relatively new class of low-cost solar cells.
They are based on a semiconductor formed between a photo-sensitized an-
ode and an electrolyte, a photoelectrochemical system. Typically a ruthe-
nium metalorganic dye (Ru-centered) used as a monolayer of light-absorbing

103
 material. The dye-sensitized solar cell depends on a mesoporous layer of
nanoparticulate titanium dioxide to greatly amplify the surface area (200-
300 m2 /gram TiO2, as compared to approximately 10 m2 /gram of flat single
crystal).

• Organic solar cells and Polymer solar cells are built from thin films (typically
100 nm) of organic semiconductors such as polymers and small-molecule
compounds like polyphenylene vinylene, copper phthalocyanine (a blue or
green organic pigment) and carbon fullerenes. Energy conversion efficiencies
achieved to date using conductive polymers are low at 6% efficiency for the
best cells to date.

• Silicon: Silicon thin-films are mainly deposited by Chemical vapor deposi-

tion (typically plasma enhanced (PE-CVD)) from silane gas and hydrogen
gas. Depending on the deposition’s parameters, this can yield: Amorphous
silicon (a-Si or a-Si:H), protocrystalline silicon or nanocrystalline silicon (nc-
Si or nc-Si:H).

• Nanocrystalline solar cells: These structures make use of some of the same
thin-film light absorbing materials but are overlain as an extremely thin
absorber on a supporting matrix of conductive polymer or mesoporous metal
oxide having a very high surface area to increase internal reflections (and
hence increase the probability of light absorption).

• CPV: Concentrating photovoltaic systems use a large area of lenses or mir-

rors to focus sunlight on a small area of photovoltaic cells. If these systems
use single or dual-axis tracking to improve performance, they may be re-
ferred to as Heliostat Concentrator Photovoltaics (HCPV).

7.4. Silicon solar cell device manufacture

Because solar cells are semiconductor devices, they share many of the same process-
ing and manufacturing techniques as other semiconductor devices such as com-
puter and memory chips. However, the stringent requirements for cleanliness and
quality control of semiconductor fabrication are a little more relaxed for solar
cells. Most large-scale commercial solar cell factories today make screen printed
poly-crystalline silicon solar cells. Single crystalline wafers which are used in the
semiconductor industry can be made into excellent high efficiency solar cells, but
they are generally considered to be too expensive for large-scale mass production.

104
 Poly-crystalline silicon wafers are made by wire-sawing block-cast silicon ingots
into very thin (180 to 350 micrometer) slices or wafers. The wafers are usually
lightly p-type doped. To make a solar cell from the wafer, a surface diffusion
of n-type dopants is performed on the front side of the wafer. This forms a p-n
junction a few hundred nanometers below the surface.
Antireflection coatings, which increase the amount of light coupled into the so-
lar cell, are typically applied next. Over the past decade, silicon nitride has gradu-
ally replaced titanium dioxide as the antireflection coating of choice because of its
excellent surface passivation qualities (i.e., it prevents carrier recombination at the
surface of the solar cell). It is typically applied in a layer several hundred nanome-
ters thick using plasma-enhanced chemical vapor deposition (PECVD) (which is
process that REC Scancell AS in Narvik uses). Some solar cells have textured
front surfaces that, like antireflection coatings, serve to increase the amount of
light coupled into the cell. Such surfaces can usually only be formed on single-
crystal silicon, though in recent years methods of forming them on multicrystalline
silicon have been developed.
The wafer is then metallized, whereby a full area metal contact is made on the
back surface, and a grid-like metal contact made up of fine "fingers" and larger
"busbars" is screen-printed onto the front surface using a silver paste. The rear
contact is also formed by screen-printing a metal paste, typically aluminium. Usu-
ally this contact covers the entire rear side of the cell, though in some cell designs
it is printed in a grid pattern. The metal electrodes will then require some kind
of heat treatment or "sintering" to make Ohmic contact with the silicon. After
the metal contacts are made, the solar cells are interconnected in series (and/or
parallel) by flat wires or metal ribbons, and assembled into modules or "solar
panels". Solar panels have a sheet of tempered glass on the front, and a polymer
encapsulation on the back. Tempered glass cannot be used with amorphous silicon
cells because of the high temperatures during the deposition process.
References: http://en.wikipedia.org/wiki/Solar_cell#Light-absorbing_materials.

105
Figure 8.1: A ”nano-tube”, 1nm in diameter and may be 1000nm long. Carbon-atoms
in a ”periodic tube”.

8. Nano - materials - and technology

Nanoelectronics, Nanotechnology, Nanoconductor, Nanocomputer, Nanobots, from
[13] are all technology based on nanoparticles which is a new functional mater-
ial with length from 1 to 100nm, where 1 nanometer is 1·10−9 meter. In recent
years, it has attracted worldwide attention in research and development of nano
materials. The immediately available applications for these materials range from
transparent skin-care products to microelectronic components and high-stress en-
gine parts.
In this near to atomic size range, the particles and dynamic properties of
surface atoms can be exploited to modify and enhance the performance of basic
raw materials (e.g. iron, aluminum and titanium). These materials have unique
electrical, optical, chemical, structural and magnetic properties with potential
applications including information storage, color imaging, bioprocessing, magnetic
refrigeration, and ferrofluids.
This type of new composite materials can be molecularly engineered. The
capabilities allow products to achieve levels of strength and durability. It can
also be use to produce designer materials with electronic, mechanical, chemical,
magnetic, and optical behavior, beyond the capability of traditional materials.
New generations of products and processes become possible.
Nanoparticles are also widely applied in such fields as astronavigation, national

106
defense, magnetic recording devices, computerization, environmental protections,
surface modification and medical/biological engineering etc.
Manufactured products are made from atoms (see [17]). The properties of
those products depend on how those atoms are arranged. If we rearrange the
atoms in coal we can make diamond. If we rearrange the atoms in sand (and add
a few other trace elements) we can make computer chips. If we rearrange the
atoms in dirt, water and air we can make potatoes.
It’s like trying to make things out of LEGO blocks with boxing gloves on
your hands. Yes, you can push the LEGO blocks into great heaps and pile them
up, but you can’t really snap them together the way you’d like. In the future,
nanotechnology will let us take off the boxing gloves. We’ll be able to snap together
the fundamental building blocks of nature easily, inexpensively and in almost any
arrangement that we desire. This will be essential if we for instance want to
continue the revolution in computer hardware.
The greatest problem of nanocrystalline materials have been to move the tech-
nology out of the laboratory and put into commercial applications by developing
means to produce quality materials in commercial volume at affordable cost. How-
ever, these difficulties have been overcome. Todays manufacturing methods have
been very crude at the molecular level. Casting, grinding, milling and even litho-
graphy move atoms in great thundering statistical herds.

8.1. Carbon Nanotubes (CNT)

Figure 8.1 gives a picture of a carbon nanotube (CNT), see ref. Aftenposten
(graphics: J. P. MÆHLEN ). One gram of these tubes costs about 5000 NOK,
and is produced by Carbon Nanotechnology Inc. (CNI) which founder is Richard
Smalley who got the Nobelprize for the invention of the technology of the nano-
tubes in 1996.
It’s worth pointing out that the word nanotechnology has become very popular
and is used to describe many types of research where the characteristic dimensions
are less than about 1,000 nanometers. For example, continued improvements in
lithography have resulted in line widths that are less than one micron: this work
is often called nanotechnology. Sub-micron lithography is clearly very valuable
but it is equally clear that lithography will not let us build semiconductor devices
in which individual dopant atoms are located at specific lattice sites. Many of the
exponentially improving trends in computer hardware capability have remained
steady for the last 50 years. There is fairly widespread confidence that these

107
trends are likely to continue for at least another ten years, but then lithography
starts to reach its fundamental limits.
Whatever we call it, nano-technology should let us:

• — Get essentially every atom in the right place.

— Make almost any structure consistent with the laws of physics and
chemistry that we can specify in atomic detail.
— Have low manufacturing costs (not greatly exceeding the cost of the
required raw materials and energy).

There are two more concepts commonly associated with nanotechnology:

• — Positional assembly.
— Self replication.

Clearly, we would be happy with any method that simultaneously achieved

the first three objectives. However, this seems difficult without using some form
of positional assembly (to get the right molecular parts in the right places) and
some form of self replication (to keep the costs down).
The need for positional assembly implies an interest in molecular robotics,
e.g., robotic devices that are molecular both in their size and precision. These
molecular scale positional devices are likely to resemble very small versions of
their everyday macroscopic counterparts. Positional assembly is frequently used
in normal macroscopic manufacturing today, and provides tremendous advantages.
The requirement for low cost creates an interest in self replicating manufac-
turing systems. These systems are able both to make copies of themselves and
to manufacture useful products. If we can design and build one such system the
manufacturing costs for more such systems and the products they make (assuming
they can make copies of themselves in some reasonably inexpensive environment)
will be very low.

8.1.1. Strength
Carbon nanotubes are one of the strongest and stiffest materials known, in terms
of tensile strength and elastic modulus respectively. Multi-walled carbon nanotube
was tested in 2000, and was found to have a tensile strength of 63 GPa. (high-
carbon steel has a tensile strength ≈ 1.2 GPa). CNTs have very high elastic-
moduli, on the order of 1 TPa. Since CNTs have a low density for a solid (1.3-1.4

108
 Figure 8.2: Picture from http://www.forskning.no.

g/cm3 ), its specific strength of up to 48462 kN·m/kg is the best of known materials,
( the specific strength of high-carbon steel is 154 kN·m/kg).
The tubes will undergo plastic deformation under excessive tensile strain,
which starts at strains of approximately 5%. CNTs are not so strong under com-
pression. Due to their hollow structure and high aspect ratio, they tend to undergo
buckling when placed under compressive, torsional or bending stress.

8.2. Nanocomposites
Materials with features on the scale of nanometers often have properties dramat-
ically different from their bulk-scale counterparts. For example, nanocrystalline
copper is five times harder than ordinary copper with its micrometer-sized crys-
talline structure. The development of such materials is currently a research area
of great interest.
Nature makes fabulous nanocomposites, and scientists are trying to emulate
such processes. The abalone shell, for example, has alternating layers of calcium
carbonate and a rubbery biopolymer; it is twice as hard and a thousand times
tougher than its components.
The study of nanocomposite materials requires a multidisciplinary approach,
involving novel chemical techniques and an understanding of physics, mathemat-
ics, material science and surface science. It is a field of broad scientific interest
with impressive technological promise.

109
Figure 8.3: (Left) This transmission electron micrograph of the cubic structure of
the new hybrid material shows pores of about 10 nanometers across. (Right) This
diagram shows molecular architecture of the flexible ceramic phase called a ”plumber’s
nightmare.” Wiesner Research Group

8.3. Flexible ceramics

Using nanoscale chemistry, researchers at Cornell developed in 2002 a new class of
hybrid materials that they describe as flexible ceramics, , see Figure 8.3 . The new
materials appear to have wide applications, from microelectronics to separating
macromolecules, such as proteins.
The Cornell researcher reasoned that the simplest way to mimic nature’s path-
ways was to use organic (or carbon-based) polymers — more particularly materials
known as diblock copolymers — that have the ability to self-assemble chemically
into nanostructures with different symmetries. If the polymer could somehow be
melded with an inorganic material — a ceramic, specifically a silica-type material —
the resulting hybrid would have a combination of properties: flexibility and struc-
ture control from the polymer and functionality from the ceramic. This, Wiesner’s
group has now achieved.
”The resulting material has properties that are not just the simple sum of
polymers plus ceramic, but maybe something quite new,” said Wiesner. Thus
far the Cornell researchers have made only small pieces of the flexible ceramic,
weighing a few grams, in petri dishes, but that is enough to test the material’s
properties. It is transparent and bendable but with considerable strength, and

110
 Figure 8.4: Flexible ceramics, from http://www.polytec.com/default.asp.

unlike pure ceramic will not shatter. In one form the hybrid material is an ion
conductor (an ion is an electrically charged atom), with great promise as highly
efficient battery electrolytes. There also is the possibility that the new material
could be used in fuel cells, he said.
Published March 21, 2002, http://www.news.cornell.edu/Chronicle/02/3.21.02/
Wiesner-ceramics.html
For example, Cotronics´ Flexible Ceramics, see Figure 8.4 are easy to use to
1650◦ C, highly efficient insulation products. They are available as:
Fiber blankets, Ceramic Papers, Moldable Sheets, Ceramic Boards, Cloths and
Sleevings, Tapes, Ropes and Fabrics. They have advantages as:
high purity, outstanding thermal insulation, low heat storage, lightweigh, high
resilienc, high mechanical and thermal shock resistance and are resistant to oxi-
dizing.

8.4. New type of High-performance Ceramic

When a racing driver brakes, the discs and linings become red-hot. These parts
are commonly made of carbon-fiber-reinforced carbon and are black at moderate
temperatures. Car manufacturers and their suppliers would dearly like to extend
the use of these special brake pads and other hard-wearing parts developed for
racing vehicles to perfectly normal family cars - if only the cost was not prohibitive.
This may now be possible using a new fiber-composite material that could almost
be referred to as a renewable ceramic. Its carbon matrix is made of carbonized
flax, hemp or wood fibers, which are then permeated with liquid silicon to form a
high-strength, wear- and temperature-resistant silicon carbide.

111
 A great deal of know-how is needed even to produce the sheets of carbon fiber,
reveals Dr. Stefan Siegel, team leader at the Fraunhofer Institute for Ceramic
Technologies and Sintered Materials IKTS. Our starting material are boards of
resin-coated plant fibers, which are produced for us by our colleagues at the Fraun-
hofer Institute for Wood Research WKI. These sheets then undergo a process
called carbonization, in which the natural-fiber composite is heated in a nitrogen
flooded furnace to temperatures of over 1000◦ C. This process has to be carefully
controlled to ensure that the shrinkage of the material - i.e. its reduction in vol-
ume - proceeds homogeneously and without distortion. All that is left in the end
is carbon; practically all other compounds present in the original material have
decomposed and evaporated. The black sheets resulting from the carbonization
of the raw natural fiber can be used, for example, as structural elements or as
an insulating lining for furnaces. Alternatively, this intermediate product can be
shaped into components using standard industrial machining techniques such as
sawing, drilling or milling.
But let us return to the ceramics. At temperatures above 1410◦ C, silicon
liquefies, and is soaked up like a sponge by the sheets of carbon (at present up to
one square meter in size) or the pre-shaped components. The silicon undergoes a
fast-acting reaction with the carbon, forming a fiber-composite ceramic material.
This stage of the process, unlike carbonization, does not alter the shape or size
of the parts. The post-processing of hard ceramic materials like silicon carbide
is relatively difficult. The new wood ceramic, by contrast, can be shaped before
it is hardened, thus presenting manufacturers who use machine tools to produce
parts subject to high mechanical and thermal stress with a simple-to-use, low-cost
alternative. (published nov. 2002).

8.5. Back to square one?

Published in : http://www.forskning.no/Artikler/2002/oktober/1034170456.0 10.
October 2002
An up-and-coming young physicist at Bell Labs in Murray Hill, New Jersey,
has been dismissed after being found guilty of 16 counts of scientific misconduct
by a review panel charged with investigating his research.
Formerly a rising star in the field of nanotechnology, Schön was renowned for
creating field-effect transistors, the backbone of modern electronics, out of tiny
molecules. His work won him numerous awards from magazines and scientific
organizations, and colleagues were beginning to tip him for a Nobel Prize.

112
 But not everything was as it seemed. Many scientists were unable to reproduce
Schön’s results. In April, a small group of physicists noticed that graphs in three
unrelated papers appeared identical down to what should have been random noise.
Bell Labs rapidly launched an independent investigation, which soon expanded to
include two dozen papers.
What they found, according to Malcolm Beasley, a professor of electrical engi-
neering at Stanford University, who chaired the panel, was that Schön substituted
whole figures from other papers, removed data points that disagreed with predic-
tions, and even used mathematical functions in place of real data points.

References:
Nano SMM:
http://mrsec.wisc.edu/Seed_shape_memory/seed_shape_memory_alloys.htm
http://en.wikipedia.org/wiki/Carbon_nanotube
Aerogel:
http://eande.lbl.gov/ECS/aerogels/aerogels.htm
http://www.ntnu.no/gemini/2002-05/16-19.htm
FGM:
http://kenwww.pi.titech.ac.jp/Current_Subjects#Topic-3
http://www.grc.nasa.gov/
http://WWW/RT2000/5000/5920arnold3.html
http://www.lucifer.com/~sean/Nano.html
http://www.zyvex.com/nano/
http://web.mit.edu/tdp/www/composite.html
CVD:
http://chiuserv.ac.nctu.edu.tw/~htchiu/cvd/home.html
http://www.ultramet.com/4.htm
nano:
http://www.aftenposten.no/nyheter/iriks/article.jhtml?articleID=199118
http://www.aftenposten.no/nyheter/iriks/article.jhtml?articleID=198172
http://www.aftenposten.no/nyheter/iriks/article.jhtml?articleID=198173
http://www.dr.dk/videnom/12nano/nano2.htm
http://www.dr.dk/orbitalen/artikler/aktnat/nanoroer.shtm
http://www.tekblad.no/show_article.asp?id=4112
http://www.pa.msu.edu/cmp/csc/nanotube.html

113
http://www.sciencesite.dtu.dk/nano/
http://www.sciam.com/nanotech/
http://www.forskning.no/Artikler/2003/juni/1055493430.67
http://www.forskning.no/Artikler/2003/september/1063028534.1
Nanocomposites:
http://www.mse.cornell.edu/materials_science_discovering/nanocomposites.html
http://www.nanocor.com/nclays/nanocomposites.htm
http://www.pnl.gov/nano/conferences/smart.html
http://www.ntnu.no/gemini/2002-05/16-19.htm
Flexible cramics:
http://www.news.cornell.edu/Chronicle/02/3.21.02/ Wiesner-ceramics.html
Solar Cell materials:
http://en.wikipedia.org/wiki/Solar_cell#Light-absorbing_materials
Misconduct in nanotechnology:
http://www.smalltimes.com/document_display.cfm?document_id=4079
http://www.nature.com/nsu/020923/020923-9.html
http://www.nanoelectronicsplanet.com/briefs/
http://nanotechweb.org/articles/news/1/9

Questions
• What is the difference between a material and a structure?
• What is a smart material?
• What kind of smart materials are there, and how do they act?
• What is a FGM?
• What is the advantage of a functional gradient material compared with tradi-
tional ”solid” materials?
• What kind of products do you think could be made from a functional gradient
material?
• What kind of products is FGM used in today? Who make them? What kind of
materials are they made of?
•What kind of Solar cell materials are there?
• What is nano-technology?
• What future is there in nano-technology?
• What is a nanotube? What is the principle of fabrication of the nanotubes?
• What kind of commercial nano-tube-based products are available in the market,
and what can they be used as?

114
• What is a flexible ceramic?

115
 Figure 9.1: Cellular structure.

9. Cellular Solids, Structures & Foams

When modern man builds large load-bearing structures, he uses dense solids; steel,
concrete, glass. When nature does the same, she generally uses cellular materials;
wood, bone, coral. There must be good reasons for it ref: (Prof. M. F. Ashby,
University of Cambridge)
Cellular solids are defined as ”...an assembly of cells with solid edges of faces,
packed together so that they fill space...”, see [5], p. 1, and see Figure 9.1. Such
materials can be found in nature, for instance cork and wood. Man-made cellular
structures will in the future replace conventional materials because of the possi-
bility to construct structures with tailored properties. Cellular structures can ei-
ther consist of two-dimensional structures, like honeycombs, or three-dimensional
structures, like foam-structures.
Honeycombs are usually thought of as cells in a hexagonal form, but they
can also be in a triangular, square or rhombic form. We can find honeycombs in
the nature, like the bee-honeycomb and as the cross-section of the bambootree.
Honeycombs are often used as a core material in sandwich-constructions. The very
good stiffness-to-weight ratio and strength-to-weight ratio and the good shock
absorption ability in addition to the possibility for man-made tailored properties
of the structures are some of the answers to why the honeycomb-structures have
become so interesting as material structures.
Foams can now also be made of almost any material, like polymers, met-
als, ceramics, glasses, and composites. The properties of a foam are related to

116
its structure and the material of which the cell walls are made. Cellular solids
are often used for thermal insulation, packaging, filtering, structural applications
or for their buoyancy property among many other applications. Recently, there
has been an increased interest in the use of cellular solids that perform more
than just one function. These multi-functional materials would possess unique
combinations of thermophysical and mechanical properties. For example, a load
bearing cellular solid could also provide for additional things such as impact/blast
absorption, thermal management, conductance of electricity and/or shielding of
electromagnetic waves, electrical power storage, filtering or impeding of fluid flow,
retardation of chemical reactions and fire or in-growth of biological tissue. The
largest obstacle in creating these materials thus far has been the inability for
many manufacturers (using foaming or other methods) to gain control over the
distribution of material/pores at the cell level. In addition, the limited choice of
base material types has slowed the introduction of cellular solids as materials of
choice for multifunctional applications.
The foam industry grows because cellular materials offer unique advantages
over traditional materials and non-cellular polymers.

9.1. Metal Foams

Metal foams are a new type of materials, that is unknown to most engineers. They
are not perfectly characterized, but generally we say that they have low densities
and new type of physical, mechanical, thermal, electrical and energy absorption.
In recent years there has been an increased interest in metal foams in the field
of material science. The stress absorbing potential is one of the most interesting
properties for the application of aluminium foam (e.g. car manufacturing). Mate-
rial scientists need to investigate the structure of metal foams in order to optimize
their deformation behavior.
Most available metal foams are based on aluminium or nickel, but there also
exist methods for foaming magnesium, lead, zinc, copper, bronze, titanium, steel
and even gold. The methods are still imperfectly controlled. Metallic foams
(metfoams) offer significant performance gains in light, stiff structures since they
have an excellent stiffness-to-weight ratio. They can also be used as core materials
in sandwich constructions since they have low density with good shear and fracture
strength. The damping capacity of metal foams is larger than that of solid metals
by up to a factor of 10.
The most promising applications for metal foams appear to be as cores for

117
 Figure 9.2: Example of aluminium foam.

light-stiff sandwich panels; as stiffeners to inhibit buckling in light shell structures;

as energy absorbing units, both inside and outside of motor vehicles and trains; as
efficient heat exchanges to cool high powered electronics (by blowing air through
the open cells of the aluminium foam, , attached to the heat source) and as light
cores for shell casting. Several industrial designers have seen potential in exploiting
the reflectivity and light-filtering of open cell foams, and the interesting textures
of those with closed cells.

9.1.1. Aluminium foam:

Also aluminium foam, see Figure 9.2, is beginning to be a large group of foam
cores. It can be difficult to distinguish between aluminium foam as core material
or as a functional gradient material, which is used in for instance car production.
The properties (mechanical, thermal et.) of foam materials is studied in the
field of cellular solids.
Blocks of foam can be used as energy absorbers in various transportation-
related applications, such as automobiles, trucks and trains and highway crash
barriers. Cymat is capable of producing aluminum foam panels up to four feet
wide and 50 feet long with thickness ranging from 1 inch to 4 inches. The densities
available are from 2.5% dense aluminum (0.0675 g/cc) to 20% dense aluminum
(0.54 g/cc).

118
9.2. Polymeric foam
Polymeric foams have important economic impact on nearly every aspect of life
today. High-density cellular plastics are used in furniture, transportation, and
building products. Low-density foams are used as shock mitigation, insulation,
and rigid packaging.
Foams can be manufactured from a variety of synthetic polymers includ-
ing polyvinyl chloride (PVC), polystyrene (PS), polyurethane (PU), polymethyl
methacrylamide (acrylic), polyetherimide (PEI) and styreneacrylonitrile (SAN).
Polyvinyl chloride PVC - Often used in the manufacture of racing boats,
as the linear construction foams are flexible, and can be heat moulded around
curves.
Closed-cell polyvinyl chloride (PVC) foams are one of the most commonly
used core materials for the construction of high performance sandwich structures.
Although strictly they are a chemical hybrid of PVC and polyurethane, they tend
to be referred to simply as ’PVC foams’.
PVC foams offer a balanced combination of static and dynamic properties and
good resistance to water absorption. They also have a large operating tempera-
ture range of typically -240◦ C to +80◦ C , and are resistant to many chemicals.
Although PVC foams are generally flammable, there are fire-retardant grades that
can be used in many fire-critical applications, such as train components. When
used as a core for sandwich construction with fiber reinforced polymer (FRP)
skins, its reasonable resistance to styrene means that it can be used safely with
polyester resins and it is therefore popular in many industries. It is normally
supplied in sheet form, either plain, or grid-scored to allow easy forming to shape.
There are two main types of PVC foam: crosslinked and uncrosslinked with
the uncrosslinked foams sometimes being referred to as ’linear’. The uncrosslinked
foams are tougher and more flexible, and are easier to heat-form around curves.
However, they have some lower mechanical properties than an equivalent density
of cross-linked PVC, and a lower resistance to elevated temperatures and styrene.
Their cross-linked counterparts are harder but more brittle and will produce a
stiffer panel, less susceptible to softening or creeping in hot climates. Typical
cross-linked PVC products include the Herex C-series of foams, Divinycell H and
HT grades and Polimex Klegecell and Termanto products.
A new generation of toughened PVC foams is now also becoming available
which trade some of the basic mechanical properties of the cross-linked PVC
foams for some of the improved toughness of the linear foams. Typical products

119
include Divincell HD grade.
Owing to the nature of the PVC/polyurethane chemistry in cross-linked PVC
foams, these materials need to be thoroughly sealed with a resin coating before
they can be safely used with low-temperature curing prepregs. Although special
heat stabilisation treatments are available for these foams, these treatments are
primarily designed to improve the dimensional stability of the foam, and reduce
the amount of gassing that is given off during elevated temperature processing.
Applications of PVC foam are in maringe, aerospace, transportation, military
and defence, and industry.
Polystyrene (PS)- Very light - around 40kg/m3 , inexpensive and easily shaped.
Although polystyrene foams are used extensively in sail and surf board man-
ufacture, where their light weight (40kg/m3), low cost and easy to sand charac-
teristics are of prime importance, they are rarely employed in high performance
component construction because of their low mechanical properties. They cannot
be used in conjunction with polyester resin systems because they will be dissolved
by the styrene present in the resin (ester-based matrices cannot be used as adhe-
sives). PS or EPS (extruded PS) is primary used as thermal insulation materials
but also in some load carrying structures. Extruded polystyrene insulation offers
a service temperature range of (-183◦ C to +74◦ C).
Polyurethane (PU) (PUR earlier)- Is very rigid and has and good mechanical
properties, but tends to deteriorate with age and become delaminated.
Polyurethane foams exhibit only moderate mechanical properties and have a
tendency for the foam surface at the resin/core interface to deteriorate with age,
leading to skin delamination. Their structural applications are therefore normally
limited to the production of formers to create frames or stringers for stiffening
components. However, polyurethane foams can be used in lightly loaded sandwich
panels, with these panels being widely used for thermal insulation. The foam also
has reasonable elevated service temperature properties (150◦ C), and good acoustic
absorption. The foam can readily be cut and machined to required shapes or
profiles.
Polymethyl methacrylamide (PMI)
For a given density, polymethyl methacrylamide (acrylic) foams such as Roha-
cell offer some of the highest overall strengths and stiffnesses of foam cores. Their
high dimensional stability also makes them unique in that they can readily be
used with conventional elevated temperature curing prepregs. However, they are
expensive, which means that their use tends to be limited to aerospace composite

120
parts such as helicopter rotor blades, and aircraft flaps. Acrylics has the high-
est specific strength and stiffness for a given density. High stability makes them
ideal for use with prepregs, but they do not fit curves well. Maximum service
temperature of PMI is 180◦ C.
Styrene acrylonitrile (SAN) co-polymer Foams
SAN foams behave in a similar way to toughened cross-linked PVC foams.
They have most of the static properties of cross-linked PVC cores, yet have much
higher elongations and toughness. They are therefore able to absorb impact levels
that would fracture both conventional and even the toughened PVC foams. How-
ever, unlike the toughened PVC’s, which use plasticisers to toughen the polymer,
the toughness properties of SAN are inherent in the polymer itself, and so do not
change appreciably with age.
SAN foams are replacing linear PVC foams in many applications since they
have much of the linear PVC’s toughness and elongation, yet have a higher tem-
perature performance and better static properties. However, they are still thermo-
formable, which helps in the manufacture of curved parts. Heat-stabilised grades
of SAN foams can also be more simply used with low-temperature curing prepregs,
since they do not have the interfering chemistry inherent in the PVC’s. Maximum
service temperature of PMI is 125◦ C.
Polyethylene (PET)- A crystalline polyethylene terephthalate (PET) foam
by applying high-expansion foam extrusion and thermoforming to PET resin.
The PET foam combines light weight and outstanding shock resistance with good
thermal insulation. Because it withstands temperatures up to 220◦ C, PET foam
is an excellent material for oven-proof food containers. And because of its superior
shock absorption and moldability, it makes an ideal packing material for multiple
products.
Other thermoplastics
As new techniques develop for the blowing of foams from thermoplastics, the
range of expanded materials of this type continues to increase. Typical is PEI
foam, an expanded polyetherimide/polyether sulphone, which combines outstand-
ing fire performance with high service temperature. PEI is expensive, but can be
used in temperatures from -194◦ C - +180◦ C, and can be used in structural, ther-
mal and fire protection applications, for instance in aircrafts and train interiors,
as it can meet some of the most stringent fire resistant specifications.

121
9.3. Refractory foams / Ceramic foam
Refractory materials are ceramics that have been designed to provide acceptable
mechanical or chemical properties while at high temperatures.
Refractory foams are a new class of open cell, low density materials which
Ultramet has developed to produce lightweight refractory structures for a variety
of aerospace and industrial applications including thermal insulation, lightweight,
precision mirrors, impact absorption, catalyst support, and metal and gas filtra-
tion. By pyrolyzing a polymer foam, a reticulated vitreous carbon (RVC) foam
is created which has substantial mechanical and thermal properties. By further
CVI (chemical vapor infiltration) processing, ceramic (e.g. silicon carbide [SiC])
and/or metal or mixed material foams can be fabricated.
These cellular materials can be simultaneously optimized for stiffness, strength,
thermal conductivity, active surface area, and gas permeability. They are ther-
mally stable, low in weight and density, chemically pure, resistant to thermal
stress and shock, and are relatively inexpensive.
June 2001: A team of scientists from Technion, Israel, has succeeded in manu-
facturing the worlds lightest ceramic foam material, implementing a unique mech-
anism its members have developed. The new material is a ceramic foam that
contains 94% to 96% air by volume, but can resist temperatures above 1700◦ C.
The material is made of aluminum oxide, a common high-temperature ceramic,
but gets its extraordinary insulating powers from the many tiny air bubbles within
the material. The foam is generated from special crystals that contain the metal
components and all the foaming ingredients. Upon heating, the crystals form a
solution. Within this solution a reaction takes place, forming polymer chains.
After the chains grow sufficiently, the solution suddenly separates into a pure
solvent and the polymer. At this point, the solvent begins to boil, forming trillions
of tiny bubbles that blow the polymer into a foam, stabilized by the polymer
chains. Subsequent heating to high temperatures leaves behind the ceramic, metal
oxide foam.
The material will be used for a wide variety of advanced applications such
as acoustic insulation, thermal insulation, adsorption of environmental pollutants
or as a platform for bio-technology applications. The estimated market for this
product is worth billions of dollars.

122
 Figure 9.3: Human bone is a hierarchical structure.

9.3.1. Carbon foam

Touchstone makes thee carbon foam product, CFoam by transforming high-sulfur
bituminous coal into a strong but lightweight, fire-resistant, and thermally insu-
lating material. A slice of Carbon foam that is exposed to a 3000◦ flame, can be
held by a hand in the other end due to its conductivity which makes that the end
remains cool enough to touch. The carbon foam material can be used as for in-
stance lightweight thermal protection in aircraft and spacecraft, as a fire blocking
material in military vehicles, as a porous electrode for lightweight energy storage,
and to provide insulation and fire protection in home construction.

9.4. Hierarchical structures (multiscale structures)

Some cellular structures are made up in a different matter than the previous ones.
The hierarchical structures (also called multiscaled structures) are structures that
consist of structures in many size scales or levels. They are also called multiscaled
structures. Many such structures can be found in nature, and humans have begun
to imitate the nature and build man-made hierarchical structures. Weight-savings
and tailored properties are some of the benefit of such structures.
Human bone (bones in general) is for instance a natural hierarchical structure,
see Figure 9.3.

123
 Figure 9.4: The Eiffel tower is also a hierarchical structure.

Eiffel-tower, on the other hand, is an example of an artifical/man-made hier-

archical structure, see Figure 9.4, is a hierarchical structure.
Most common materials or structures have a Poissons ratio 0.5 > ν > 0
(approximately 0.5 for rubbers and for soft biological tissues, 0.45 for lead, 0.33
for aluminum, 0.27 for common steels, 0.1 to 0.4 for cellular solids such as typical
polymer foams, and nearly zero for cork). In an isotropic material (a material
which does not have a preferred orientation) the allowable range of Poisson’s ratio
is from -1.0 to +0.5. Hierarchical structures and foams have also been made with
negative Poissons ratio (ν < 0), and Poisson’s ratio > 1, which give exciting
respond to loads. The applications for such structures/materials are not widely
known, but will certainly increase in future when the fabrication of such structures
will be easier. Rod Lakes (at: http://silver.neep.wisc.edu/~lakes/sci87.html) is
maybe the person who has done most in this field.

9.5. Biomaterials
A biomaterial is any material, natural or man-made, that comprises whole or part
of a living structure or biomedical device which performs, augments, or replaces
a natural function.
References:

124
Cellular structures:
http://ansatte.hin.no/dl//homo.html
http://web.mit.edu/dmse/csg/Research.htm
http://silver.neep.wisc.edu/~lakes/
Metal foam:
http://www.metalfoam.net/
http://www.porvair.com/fuelcell/metalfoam.htm
http://www.porvair.com/fuelcell/
http://www.ultramet.com/foamtech.htm
http://www.cymat.com/
http://www.npl.co.uk/npl/cmmt/metal_foams/index.html
http://www.grantadesign.com/solutions/metalfoams.htm

Polymer foam:
http://www.vitec.org.uk/
http://www.foamandform.com/seminars/foam/
Ceramic foams:
http://www.emgroup.nl
http://www.solgel.com/articles/june01/news06_101.htm
Refractory foams:
http://www.ultramet.com/
http://voyager.cet.edu/iss/techcheck/techcheck3/wvproduct.html
General:
http://www.npl.co.uk/npl/cmmt/metal_foams/applications.html
Hierarchical structures
http://silver.neep.wisc.edu/~lakes/Hierarch.html
Negative Poissons ratio materials
http://silver.neep.wisc.edu/~lakes/sci87.html
Biomaterials:
http://www.engr.sjsu.edu/WofMatE/Biomaterials.htm
Questions
• What is the definition of ”cellular solids”?
• What types of foams are there, and what products are they used in?

125
• What is the advantages of foam materials compared to conventional solid ma-
terials?
• What are the applications of foam materials?
• What is a refractory material/structure?
• What is hierarchical structures?
• Find two examples of man-made hierarchical structures and two examples of
natural hierarchical structures.
• Does foam with negative Poissons ratio exist? How does such a material act?
• Does foam with Poissons ratio >1 exist?

126
Part II
Sandwich Constructions
Composite materials used today are often in the form of a sandwich construction,
see Figure 9.5.A sandwich panel is built up by two thin skins, also called the

Face sheet

Honeycomb core
(metal, composite
or paper)

Adhesive

Face sheet

Figure 9.5: A sandwich construction.

facings, separated by a lightweight core. The core helps to increase the moment of
inertia such that the structure becomes efficient for resisting bending and buckling
loads. This is why sandwich panels are being used in applications where weight-
saving is critical, for instance in aircraft and in portable structures.
Sandwich designs are also used by nature itself for instance in a human skull
and in plants. While nature often uses the same material in the facings as in
the core and only vary the density, man-made sandwich panels usually consist of
different materials, or even structures, in the facings and in the core.A sandwich
construction is built up by different face- and core materials and forms, see Figure
9.6 and [36, p.18].

127
 Figure 9.6: Sandwich materials and forms.

10. Why use sandwich constructions?

The mechanical behavior of a sandwich panel depends on the properties of the
face and the core materials and on its geometry.
The Figure 10.1 shows how the sandwich design works, the so-called sandwich-
effect.
In Figure 10.2 we see the sandwich-effect in and aluminium-sandwich compared
to solid aluminium plates of different thickness.

128
 Figure 10.1: The sandwich effect.

Figure 10.2: Aluminium foam used as core material compared with aluminium plate.

129
10.1. Face materials
The face material in a sandwich panel can be made of almost any material that
can be formed into thin sheets. The properties we seek for in a face material are:
• high stiffness, which gives high flexural rigidity
• high tensile and compressive strength
• impact resistance
• surface finish
• environmental resistance (chemical, UV, heat etc.)
• wear resistance
The most usual type of material used as facings are listed in the Table with a
summary of face materials.

Material
Metals: Mild steel
Stainless steel
Aluminium Alloy
Titanium Alloy
Wood: Pine
Plywood
Unidirectional fibre composites Carbon/Epoxy , (10.1)
(vf = 0.6 − 0.7) Glass/Epoxy
Kevlar/Epoxy
Bi-directional fibre composites Kevlar/Polyester
(vf = 0.3 − 0.4) Glass weave/Polyester
Glass WR /Polyester
Random fibres Glass CSM
(vf = 0.15 − 0.25) SMC
where WR=(woven roving), CSM= chopped strand mat, SMC = sheet mould-
ing compound and vf is the volume fraction of fibers.
The most used group of face-materials is the fibre composites since they have
a similar or even higher strength properties than metals, and are much easier to
fabricate. Also, the possibility to tailor the face materials because the anisotropic
behavior of the fibres offer a very interesting addition. In this way fibres can be
placed in the direction where the loadcarrying is most important.

130
 Figure 10.3: Core exposed to shear.

10.2. Core materials and structures

The essential property of any core material is that it increases the thickness of the
laminate, without causing a great weight increase (engineering theory shows that
the flexural stiffness of any panel is proportional to the cube of its thickness). The
purpose of a core in a composite laminate is therefore to increase the laminate’s
stiffness by effectively ’thickening’ it with a low-density core material. This can
provide a dramatic increase in stiffness for very little additional weight.
Figure 10.3 shows a cored laminate under a bending load. Here, the sandwich
laminate can be likened to an I-beam, in which the laminate skins act as the
I-beam flange, and the core materials act as the beam’s shear web. In this mode
of loading it can be seen that the upper skin is put into compression, the lower
skin into tension and the core into shear. It therefore follows that one of the most
important properties of a core is its shear strength and stiffness.
In addition, particularly when using lightweight, thin laminate skins, the core
must be capable of taking a compressive loading without premature failure. This
helps to prevent the thin skins from wrinkling, and failing in a buckling mode.
The properties of primary interest for the core:
• Low density (add little weight to the total weight of the sandwich)
• Shear modulus (prevents wrinkling)
• Shear strength (prevents wrinkling)
• Stiffness perpendicular to the faces (prevents decrease in core thickness and
therefore a rapid decrease in the flexural rigidity)
• Thermal insulation
• Acoustic insulation
There are four main groups of core material used,

131
 Figure 10.4: Honeycomb core.

• Foams
• Honeycombs
• Corrugated
• Wood

10.2.1. Foam Cores

Foams are one of the most common forms of core material. They can be manu-
factured from a variety of synthetic polymers including polyvinyl chloride (PVC),
polystyrene (PS), polyurethane (PU), polymethyl methacrylamide (acrylic), poly-
etherimide (PEI) and styreneacrylonitrile (SAN). They can be supplied in densi-
ties ranging from less than 30kg/m3 to more than 300kg/m3 , although the most
used densities for composite structures range from 40 to 200 kg/m3 . They are
also available in a variety of thicknesses, typically from 5mm to 50mm.

10.2.2. Honeycomb Cores

Honeycomb cores are available in a variety of materials for sandwich structures,
which ranges from paper and card for low strength and stiffness, low load appli-
cations (such as domestic internal doors) to high strength and stiffness, extremely
lightweight components for aircraft structures. Honeycombs can be processed into
both flat and curved composite structures, and can be made to conform to curves
without excessive mechanical force or heating.
Figure 10.4 shows the usual honeycomb shape. The cells can be triangular,
square or hexagonal cells which also can be filled with a rigid foam in order to

132
provide a greater bond area for the skins, increases the mechanical properties of
the core by stabilizing the cell walls and increases thermal and acoustic insulation
properties. The following materials are commonly used in honeycomb structures.

• Aluminium - Has been used since 1950, several alloys can be used, but in
comparison it is old and heavy.

• Glass fibre reinforced plastic - Has a high temperature resistance and good
insulating properties, but is denser than other materials.

• Kraftpaper honeycombs - impregnated paper with resin to make it water

resistant. Good strength at low cost.

• Nomex honeycomb which is made from Nomex paper - a form of paper based
on KevlarTM (Aramid fibre), rather than cellulose fibres. High strength and
toughness with a low density makes it the most widely used honeycomb
core. The initial paper honeycomb is usually dipped in a phenolic resin to
produce a honeycomb core with high strength and very good fire resistance.
It is widely used for lightweight interior panels for aircraft in conjunction
with phenolic resins in the skins. Nomex honeycomb is becoming increas-
ingly used in high-performance non aerospace components due to its high
mechanical properties, low density and good long-term stability.

Figure 10.5 shows the shear strength and compressive strength of some of the
core materials described, plotted against their densities. Figure 10.6 shows the
prices of some core materials.

Honeycombs can be made with several different cell shapes:

Hexagonal
The most common shape is the one shown at Figure 10.4, but this can only
be used in flat components.
Overexpanded
The honeycomb in Figure 10.7 is over expanded, so that the cells are rectan-
gular. This gives better properties in the web direction, but worse in the other, it
can therefore be curved in the ribbon direction only. This shape is used for single
curvature components.

133
Figure 10.5: Compressive strength and shear strength of some of the core materials
plotted against their densities.

Figure 10.6: Comparative prices of some core materials.

134
 Figure 10.7: Overexpanded honeycomb structure.

Negative Poisson’s Ratio

Honeycombs can be made with negative Poissons Ratio, when the cell walls
are inverted as in Figure 10.8.

Figure 10.8: Honeycombs with negativ Poisson’s ratio.

10.2.3. Corrugated Cores

A corrugated core is shown in Figure 10.9.

135
 Figure 10.9: Corrugated core.

10.2.4. Wood Cores

Wood can be described as ’nature’s honeycomb’ because on a microscopic scale
you find that it consists of closed-cell structure. It has a similar structure to
that of a hexagonal honeycomb, and consequently good mechanical properties.
When used in a sandwich structure with the grain running perpendicular to the
plane of the skins, the resulting component shows properties similar to those made
with man-made honeycombs. However, despite various chemical treatments being
available, all wood cores are susceptible to moisture attack and will rot if not well
surrounded by laminate or resin. Wood is only used in large projects, as it has a
relatively high density, of at least 100kg/m3 .

Balsa
The most commonly used wood core is end-grain balsa. Balsa wood cores first
appeared in the 1940’s in flying boat hulls, which were aluminium skinned and
balsa-cored to withstand the repeated impact of landing on water. This perfor-
mance led the marine industry to begin using end-grain balsa as a core material
in FRP construction. Apart from its high compressive properties, its advantages
include being a good thermal insulator offering good acoustic absorption. The
material will not deform when heated and acts as an insulating and ablative layer
in a fire, with the core charring slowly, allowing the non-exposed skin to remain
structurally sound. It also offers positive flotation and is easily worked with simple
tools and equipment.
Balsa core is available as contoured end-grain sheets 3 to 50mm thick on a
backing fabric, and rigid end-grain sheets up to 100mm thick. These sheets can
be provided ready resin-coated for vacuum-bagging, prepreg or pressure-based
manufacturing processes such as RTM. One of the disadvantages of balsa is its

136
high minimum density, with 100kg/m3 being a typical minimum. This problem
is exacerbated by the fact that balsa can absorb large quantities of resin during
lamination, although pre-sealing the foam can reduce this. Its use is therefore
normally restricted to projects where optimum weight saving is not required or in
locally highly stressed areas.
Balsa was the first material used as cores in load bearing sandwich structures
and is still often used as a core material.

Cedar
Another wood that is used sometimes as a core material is cedar. In marine
construction it is often the material used as the ’core’ in strip-plank construction,
with a composite skin on each side and the grain of the cedar running parallel to
the laminate faces. The cedar fibres run along the length of the boat giving fore
and aft stiffness while the fibres in the FRP skins are laid at ±45◦ giving torsional
rigidity, and protecting the wood.

10.3. Adhesives
Bonding of sandwich construction involve bonding of two very dissimilar con-
stituents, one solid and one softer cellular component, and the requirements con-
cerning bonding are therefore somewhat different than normal use. The adhesive
must be stronger than the tensile strength of the core. Some of the requirements
of the adhesives are:
Surface preparation
The core and the face material have to be prepared before bonding, which
involves mechanically or chemically cleaning and sometimes priming.
Solvents
Core materials are often very sensitive to certain solvents. For instance: Poly-
styrene foams are sensitive to styrene (polyester and vinylester contains styrene),
while epoxies and polyurethanes may be used. Similar combinations needs to be
investigated before bonding components.
Curing vapors
When curing, some adhesives (as phenolics) give off vapor when curing, which
can give rise to several bonding problems.
Bonding pressure

137
 When pressure is needed to prevent pores to appear, be careful so that the
core will not fail due to the compression.
Adhesive viscosity
The adhesive must have exactly the right combination of surface wetting and
flow. In the case of foam or balsa core, the viscosity should be low enough to
enable the adhesive to fill the surface cells properly and leave as little as possible
trapped air. But the viscosity must not be too low,the adhesive could be squeezed
out leaving too thin bonding line.
Bond thickness
If the bond is too thick, it adds extra unnecessary weight to the part. If it is
too thin, bonding will not be one properly.
Strength
The bond must be able to transfer the design loads, which means it must have
the desired tensile and shear strength, at the temperatures that might occur.
Thermal stresses
A frequent cause of debonding failures are thermal stresses. If for instance one
side is heated from sunlight it will deform due to thermal expansion. Most core
materials are very good insulators, and therefor it will be a very high thermal
gradient over the bond line. This lead to very high shear stresses in the bond
which may lead to debonding. In such environment, very ductile adhesives should
be chosen (high strain to failure).
Toughness
Toughened adhesives which resist cracks better (improved impact resistance)
are on the market. They are ordinary resins which have elastomer particles added.
Viscoelastic properties
Highly viscoelastic adhesives may be advantageous for example where there
are high thermal gradients.
Curing shrinkage
As much as 7% decrease in volume can an adhesive (as polyesters) shrink when
its curing. This leads to high interface (bond) shear stresses and may decrease
the strength of adhesive joints.
Curing exotherm
Most adhesives exhibit an exotherm (curing process gives off heat) curing.
This is seldom a problem in thin bondings spread over a large area.

138
 Different types of adhesives are for instance:
• Epoxy resins
• Modified epoxies
• Phenolics
• Polyurethanes Urethane acrylates
• Polyester and vinylester resins

References:
Alulight products:
http://www.alulight.com/en/products/products.html
Composites:
http://www.users.globalnet.co.uk/~weeks/
Composite%20Materials.htm
http://www.netcomposites.com/education.asp
http://www.baltek.com/
Fibres
http:// www.netcomposites.com/education.asp?sequence=45
Foam cores:
http://www.netcomposites.com/education.asp
http://www.polymer-age.co.uk/techlink.htm
http://www.hexcelcomposites.com
Adhesives:
http://www.hexcelcomposites.com/products/honeycomb/
sand_design_tech/hsdt_p04.html
http://www.hexcelcomposites.com/products/
Questions
•Why are sandwich constructions used?
•What kind of face and core materials are used in sandwich constructions?
•What is the essential property of any core material
•Is End Grain Balsa used as core material in sandwich construction? If yes,
what kind of products, and what are the advantages of this product?
(http://www.baltek.com/)
•What requirements must be taken into consideration when it comes to bonding
sandwich-constructions with adhesives?

139
11. Design of sandwich constructions
When we use sandwich panels in different applications we have to know the me-
chanical properties of the face and core materials and the geometry of the panel.
Often we formulate the design of a sandwich panel as an optimization problem
where the goal is to minimize the weight of the panel that meet the constraints
on stiffness and strength desired. With respect to the core and skin thickness, or
the materials, or the density of the core, the optimization can be carried out.

11.1. Design of sandwich beams

Design of sandwich beams are the basis for understanding more complicated panels
and constructions. Beams are defines as a long, slender, one dimensional structural
element that carries load primarily in bending (flexure). In classical engineering
beam theory where the cross-section of the beam is a homogenous material the
transverse shear deformations are neglected. In sandwich beam theory we cannot
neglect the transverse shear deformations.
An example of a sandwich beam can be seen in Figure 11.1. Here, the beam
has a given point load P at the middle of the length.l We use the subscripts ”f”,

Figure 11.1: A sandwich beam.

and ”c” which refer to the facings, and core, respectively.

140
 Figure 11.2: A curved beam.

11.2. Preliminaries
Let us start with recalling the problem of a straight beam subjected to a constant
bending moment giving the beam a curvature 1/Rx . The length of the curve, lc at
the neutral axis (N.A.), that has radius Rx and angle α, see Figure 11.2, is found
by the formula
lc = Rx α.
The strain εx in a new curveline at a distance z from the N.A. (where εx = 0) can
be found as follows:
(new curve length)-(old curve length) (Rx + z) α − Rx α
εx = =
(old curve length) Rx α
Rx α + zα − Rx α z
= = .
Rx α Rx
The stress σ x in the beam is then given by
z
σ x = Eεx = E .
Rx
The applied bending moment dMx , with respect to the x-axis, caused by the force
dF acting on a small element, see Figure 11.3, of area dA = dydz at a point x is

dMx = zdF = zσ x dA = zσ x dydz .

| {z } | {z }
dF dA

141
 Figure 11.3: A small element dA.

The total moment at the point x of the beam with height h and depth b is
X Z h/2 Z b/2 Z h/2 Z b/2
Mx = dMx = dMx = zσ x dydz
−h/2 −b/2 −h/2 −b/2
Z h/2 µ
Z b/2 ¶ Z h/2 Z b/2
z z2
= z E dydz = E dydz
−h/2 −b/2 Rx −h/2 −b/2 Rx
Z h/2 Z b/2 Z h/2 Z b/2
1 1
= Ez 2 dydz = Ez 2 dydz
Rx −h/2 −b/2 Rx −h/2 −b/2
Z h/2
1 1 1
= bEz 2 dz = (EI)eq = D,
Rx −h/2 Rx Rx

where Z h/2
(EI)eq = b Ez 2 dz (11.1)
−h/2

(often also denoted D) is called the equivalent flexural rigidity, also called the
bending stiffness. The general expression for the strain will then be
µ ¶ Ã !
z 1 Mx Mx
εx = = z= z= z. (11.2)
Rx Rx D (EI)eq

The strain will vary linearly with z over the cross-section. The basic equations
for the sandwich beam are now established, and this makes it possible to find the
cross-sectional properties and stresses in such a beam.

142
11.3. The Flexural Rigidity and Shear Rigidity
The theory for engineering stresses in beams is easily adapted to sandwich beams
if we modify it slightly.

Flexural rigidity
In an ordinary beam the flexural rigidity, denoted EI, where E is the modulus
of elasticity (Young’s modulus) and I is the second moment of area. The equivalent
flexural rigidity of a cross-section of a sandwich (see Allan 1969) is the sum of
flexural rigidities of the different parts in the sandwich and can be found from
(11.1) and is
bt3f btf d2 bt3
(EI)eq = Ef + Ef + Ec c = 2 (EI)f + (EI)o + (EI)c , (11.3)
6 2 12
where d (= tf + tc ) (see Figure 11.1) is the distance between the centroids of the
faces. The first term, 2 (EI)f , represents the flexural rigidity of the faces alone
when they are bending about their own neutral axes, and the third term, (EI)c ,
represents the flexural rigidity of the core. The first and the third term are small
compared to the second term, (EI)o , which corresponds to the stiffness of the
faces associated with bending about the centroid axis of the whole sandwich.

The faces in a sandwich are thin compared to the core (tf << tc ), hence the
first term in (11.3) is less than 1% of the second term if
µ ¶2
d d
3 > 100 or > 5.77 (11.4)
tf tf
because ⎛ ⎞
bt3
Ef 6f
⎝ ⎠ 100 < 1.
bt d2
Ef f2
The first term can therefor be neglected and (11.3) is the reduced to
btf d2 bt3c
(EI)eq = Ef + Ec = (EI)o + (EI)c .
2 12
If the core is weak, then Ec << Ef , the third term of (11.3) is less than 1% of
the second term if
6Ef tf d2
> 100, (11.5)
Ec t3c

143
because à 3
!
Ec bt12c
btf d2
100 < 1.
Ef 2

The third term and can therefore be neglected, and (11.3) is reduced to the ap-
proximated simple formula

Ef btf d2
(EI)eq = = (EI)o . (11.6)
2

Shear rigidity
The shear rigidity AG (shear stiffness) for a beam with a homogeneous cross-
section is given by
bhG
AG = ,
k
where h is the height, G is the shear modulus, and k is a shear factor, which for
rectangular homogenous cross-sections equals 1.2.
The equivalent shear rigidity (shear stiffness) for a sandwich beam when Ec <<
Ef and tf << tc is given as
bd2 Gc
(AG)eq = ,
tc
where Gc is the shear modulus of the core. We will come back to this later.

11.4. Tensile and Compressive Stresses

In a sandwich the faces carry bending moments as tensile and compressive stresses,
while the core carries the transverse forces as shear stresses as illustrated in Figure
11.4. The distribution of stresses and shear stresses in a sandwich is shown in
Figure 11.4.

11.4.1. Due to bending (transversal loading)

From (11.2) the tensile and compressive stresses (also called direct stresses) in the
sandwich due to bending shown in Figure 11.5 is given by
à !
Mx
σ = εE = z E. (11.7)
(EI)eq

144
 Figure 11.4: Stresses at different approximation levels in a sandwich.

Thus, the stress in the face σ f of the sandwich due to bending can be written as

Mx zEf tc tc
σf = for < |z| < + tf . (11.8)
(EI)eq 2 2

Figure 11.5: A sandwich beam exposed to (transversal) bending load.

145
If Mx is positive then the maximum stress in the face appears when z = tc /2 + tf :
¡ ¢ ¡ ¢
Mx zEf Mx t2c + tf Ef Ef Mx t2c + tf
σ f max = = = ³ ´
(EI)eq (EI)eq Ef btf d2
2
¡ tc ¢
2Mx 2 + tf Mx tc 2Mx
= 2
= +
btf d btf d2 bd2

while the minimum stress (in the same face) appears when z = tc /2:
¡ ¢ ¡ ¢
Mx zEf Mx t2c Ef Ef Mx t2c M t
σ f min = = = ³ 2
´ = x 2c .
(EI)eq (EI)eq E bt
f f d btf d
2

Hence
Mx tc Mx tc 2Mx
2
≤ σf ≤ +
btf d btf d2 bd2
in that face. When tf << tc this shows that we can use the following approxima-
tive formula:
Mx tc Mx d Mx
σf ≈ ≈ = (11.9)
btf d2 btf d2 btf d
(in order to see this we just use that
2Mx
bd2 tf
Mx tc
=2 ≈ 0).
btf d2
tc

(For the other face we obtain similarly that

Mx tc Mx d Mx
σf ≈ − 2
≈ − 2
=− .)
btf d btf d btf d

We obtain the opposite signs if Mx is negative.

The stress in the core σ c of the sandwich due to bending is

Mx zEc tc
σc = for |z| < . (11.10)
(EI)eq 2

146
 Figure 11.6: A sandwich beam exposed to (in-plane) axial load F.

If Mx is positive then the maximum stress in the core appears when z = tc /2, i.e.
¡ ¢ ¡ ¢
Mx zEc Mx t2c Ec Mx t2c Ec
σcmax = = = ³ ´
(EI)eq (EI)eq Ef btf d2
2
¡ tc ¢
2Mx 2 Ec Mx tc Ec
= 2
=
Ef btf d Ef btf d2
while the minimum stress in the core appears when z = 0,
Mx zEc Mx (0) Ec
σ cmin = = = 0.
(EI)eq (EI)eq

Hence
Mx tc Ec Mx tc Ec Mx Ec
0 ≤ σc ≤ 2
≤ 2
= .
Ef btf d Ef btf tc Ef btf tc
Thus when Ec << Ef we can use the following approximative formula:
Mx tc Ec Mx dEc Mx Ec
σc = ≤ = ≈ 0. (11.11)
Ef btf d2 Ef btf d2 Ef btf d
The direct stress will vary linearly within each material constituent, but there
will be a jump in the stress at the interface between the face and the core, see
Figure 11.4.

11.4.2. Due to in-plane loading (axial)

The direct strain and stress in a sandwich due to in-plane loading shown in Figure
11.6 is
Fx
εx0 = ,
b (2Ef tf + Ec tc )

147
where Fx is the axial load (normal force) and εx0 is the strain at the neutral axis
(N.A.) if the two facings are of equal thickness (in order to see this, we note that
the strain εx0 will be the same in the core and faces so that
Fx = Ef εx0 2tf b + Ec εx0 tc b = εx0 b (2Ef tf + Ec tc )).
| {z }|{z} | {z }|{z}
σf Af σc Ac

The stresses will be

σ f = Ef εx0 and σ c = Ec εx0 .
The stresses and strains due to bending and in-plane loading can be superim-
posed.

11.5. Shear stresses

Consider a small
P i volume element with sidelengths dx, dy and dz see Figure 11.7.
Using that Fx = 0 where Fxi are the forces acting on the cell walls i in the
x-direction, we obtain
X ∂σ x
Fxi = −σ x dydz + (σ x + dx)dydz
∂x
µ ¶
∂τ xy
−τ xy dxdz + τ xy + dy dxdz
∂y
∂τ xz
−τ xz dxdy + (τ xz dxdy + dz)dydx
∂z
∂σ x ∂τ xy ∂τ xz
= dxdydz + dydxdz + dzdydx
∂x ∂y ∂z
µ ¶
∂σ x ∂τ xy ∂τ xz
= + + dxdydz = 0.
∂x ∂y ∂z
Hence,
∂σ x ∂τ xy ∂τ xz
+ + = 0. (11.12)
∂x ∂y ∂z
P i P i
Similarly, by using that Fy = 0 and Fz = 0, we obtain the other two
equations for equilibrium:
X ∂τ yx ∂σ y ∂τ yz
Fy = + + =0 (11.13)
∂x ∂y ∂z

148
 Figure 11.7: A small element.

and X ∂τ zx ∂τ zy ∂σ z
Fz = + + = 0. (11.14)
∂x ∂y ∂z
It can also be verified that

τ xy = τ yx , τ xz = τ zx , τ yz = τ zy . (11.15)

Equations (11.12), (11.13), (11.14), and (11.15) are called the equilibrium equa-
tions.

We assume that the shear stress in (11.12) will not vary with y, and therefore
∂τ xy /∂y = 0 (in fact we assume τ xy = 0 in all points). Thus we get that
X ∂σ x ∂τ xz
Fxi = + =0
∂x ∂z
and the shear stress τ xz can therefore be found as follows:
∂σ x ∂τ xz
=− .
∂x ∂z
This gives
Z (d+tf )/2 Z (d+tf )/2
∂σ x ∂τ xz
dz = − dz = −(τ xz ((d + tf )/2) − τ xz (z)) = τ xz (z),
z ∂x z ∂z | {z }
=0

(when using the fact that τ xz at (d + tf )/2 = 0) i.e.

Z (d+tf )/2
∂σ x
τ xz (z) = dz.
z ∂x

149
From equation (11.7) Ã !
Mx
σ = εE = z E
(EI)eq
and the fact that
dMx
= Tx
dx
where Tx is the shear force. We obtain that the shear stress τ xz (z)
³ ´
Z (d+tf )/2 d Mx zE Z (d+tf )/2
(EI)e q Ez dMx
τ xz (z) = dz = dz =
z dx z (EI)eq dx
Z (d+tf )/2 Z (d+tf )/2
zE Tx Tx
= Tx dz = zEdz = B(z),
z (EI)eq (EI)eq z (EI)eq

where B(z) is the first moment of area

Z (d+tf )/2
B(z) = zEdz.
z

We observe that for |z| < tc /2

Z (d+tf )/2 Z tc /2 Z (d+tf )/2
B(z) = zEdz = zEc dz + zEf dz
z z tc /2
∙ ¸t /2 ∙ ¸(d+tf )/2
1 2 c 1 2
= Ec z + Ef z
2 z 2 t /2
"µ ¶ # "cµ ¶2 µ ¶2 #
2
Ec tc E f d + tf tc
= − z2 + −
2 2 2 2 2
∙ 2 ¸ " #
Ec tc 2 Ef (d + d − tc )2 t2c
= −z + −
2 4 2 4 4
∙ 2 ¸ ∙ 2 ¸
Ec tc 2 Ef Ec tc 2 Ef dtf
= −z + d[d − tc ] = −z + .
2 4 2 | {z } 2 4 2
=tf

Thus, the shear stresses in the core of the sandwich is

µ µ ¶¶
Tx Ef tf d Ec t2c 2 tc
τc = + −z for |z| < . (11.16)
(EI)eq 2 2 4 2

150
 For tc /2 ≤ −z ≤ tc /2 + tf (in the upper face)
Z (d+tf )/2 Z −tc /2 Z tc /2 Z (d+tf )/2
B(z) = zEdz = zEf dz + zEc dz + zEf dz
z z −tc /2 tc /2
| {z }
=0
∙ ¸−tc /2 ∙ ¸(d+tf )/2
1 2 1
= Ef z + Ef z 2
2 z 2 tc /2
"µ ¶ # "µ ¶2 µ ¶2 #
2
1 tc 1 d + tf tc
= Ef − z 2 + Ef −
2 2 2 2 2
∙ 2 ¸ "µ ¶2 # õ ¶2 !
2
Ef tc Ef d + tf t Ef d + tf
= − z2 + − c = − z2
2 4 2 2 4 2 2
õ ¶2 ! µ ¶
Ef (tf + tc ) + tf 2 Ef 1 2 2 2
= −z = t + tf tc + tf − z .
2 2 2 4 c

For tc /2 ≤ z ≤ tc /2 + tf (in the lower face)

Z Z "µ ¶2 #
(d+tf )/2 (d+tf )/2
1 d + tf
B(z) = zEdz = zEf dz = Ef − z2
z z 2 2
õ ¶2 ! µ ¶
Ef (tf + tc ) + tf 2 Ef 1 2 2 2
= −z = t + tf tc + tf − z
2 2 2 4 c

(i.e. the same as for the upper face). Hence, the shear stress in the facings of the
sandwich is
µ ¶
Tx Ef 1 2 2 2
τf = t + tf tc + tf − z for tc /2 ≤ |z| ≤ tc /2 + tf . (11.17)
2 (EI)eq 4 c

Assume that Tx is positive (otherwise max is replaced by min and vice versa).

Then the maximum shear stress in the core appears when the expression
(t2c /4− z 2 ) in (11.16) is maximized, i.e. when z = 0 (the neutral axis N.A.),
and is found in the following way:
µ µ ¶¶ µ µ ¶¶
Tx Ef tf d Ec t2c 2 Tx Ef tf d Ec t2c
τ c max = + −z = + −0
(EI)eq 2 2 4 (EI)eq 2 2 4

151
 µ ¶ µ ¶
Tx Ef tf d Ec t2c Tx Ef tf d Ec t2c
= + =³ ´ +
(EI)eq 2 8 Ef btf d2 2 8
2
µ ¶
Tx Ec t2c
= Ef tf d + . (11.18)
(Ef btf d2 ) 4

The minimum shear stress in the core appears when the expression (t2c /4 − z 2 )
in (11.16) is minimized, i.e. when |z| = tc /2 (the face/core interface), i.e.
µ µ ¶¶ µ µ ¶¶
Tx Ef tf d Ec t2c 2 Tx Ef tf d Ec t2c t2c
τ c min = + −z = + −
(EI)eq 2 2 4 (EI)eq 2 2 4 4
µ ¶ µ ¶
Tx Ef tf d Tx Ef tf d Tx
= =³ ´ = . (11.19)
(EI)eq 2 Ef btf d2 2 bd
2

³ 2 The maximum shear ´ stress in the face appears when the expression
tc 2 2
4
+ tc tf + tf − z in (11.17) is maximized, i.e. when z = tc /2 (the face/core
interface) such that
µ 2 ¶ Ã µ ¶2 !
2
Tx Ef tc Tx Ef tc tc
τ f max = + tc tf + t2f − z 2 = + tc tf + t2f − =
(EI)eq 2 4 (EI)eq 2 4 2
µ ¶
Tx Ef t2c t2 Tx Ef ¡ ¢ Tx Ef ¡ ¢
= + tc tf + t2f − c = tc tf + t2f = ³ 2
´ tc tf + t2f =
(EI)eq 2 4 4 2 (EI)eq E bt d
2 f 2f
Tx Ef Tx Tx
= (tc + tf ) = (tc + tf ) = (11.20)
Ef bd2 bd2 bd
(this also shows that τ c min = τ f max ).

³ 2 The minimum shear ´ stress in the face appears when the expression
tc 2 2
4
+ tc tf + tf − z in (11.17) is minimized, i.e. when |z| = tc /2 + tf

µ ¶ Ã µ ¶2 !
Tx Ef t2c Tx Ef t2c tc
τ f min = + tc tf + t2f − z 2 = + tc tf + t2f − + tf
(EI)eq 2 4 (EI)eq 2 4 2

152
 µ µ ¶¶
Tx Ef t2c 2 2 1 2 Tx Ef
= + tc tf + tf − tf + tf tc + tc = (0) = 0.
(EI)eq 2 4 4 2 (EI)eq
The deviation between the maximum and minimum shear stress in the core
from (11.18) and (11.19) will be less than 1 percent, i.e.
τ c max − τ c min 1
< ,
τ c min 100
if

µ Ã !
³ ´¶ ³ ´
|Tx | Ec t2c |Tx | ¶ Ef tf d
Ef tf d + − µ
Ef btf d2
(Ef btf d2 ) 4 2
τ c max − τ c min 2 1
= ³ ´ < .
τ c min µ |Tx | ¶
Ef tf d 100
Ef btf d2 2
2

This means that

Ec t2c 1
< ,
4Ef tf d 100
or
4Ef tf d
100 <
Ec t2c
which is often satisfied for sandwich beams because we often have that the core is
weak i.e. Ec << Ef and the faces are thin i.e. tf << tc . Thus we often use τ c min
as an approximation for the shear stress in the core (see Figure 11.4).

11.5.1. Summary of approximations

If we assume weak core i.e. Ec << Ef and thin facings i.e. tf << tc, the stresses
in the face and core in the sandwich are
Mx
σ f,max = (11.21)
btf d

from (11.9) since tc ≈ d, and

σ cmax ≈ 0 (11.22)
from (11.11).
As an approximation for the shear stress in the core τ c min we can use

153
 Tx
τc = (11.23)
bd
from (11.20).

154
11.6. Sandwich design: stiffness, strength and weight
The deflection, δ of a sandwich beam in general is the sum of the bending and

Figure 11.8: Deflection of a cantilever sandwich beam with an end point load..

shear components, see Figure 11.8 (subscripts ”b” and ”s” denotes bending and
shear, respectively),

P L3 PL
δ = δb + δs = +
B1 (EI)eq B2 (AG)eq

where B1 and B2 are constants which depend on the geometry of the loading and
the boundary conditions (type of support). Moreover, the maximum moment Mx
and the maximum shear (transverse) force Tx are given by

PL P
Mx = , Tx = ,
B3 B4

155
where B3 and B4 also are constants which depend on the geometry of the loading
and the boundary conditions. In Table 11.24 you can find the constans for several
loading conditions and boundary conditions.

Mode of loading,
B1 B2 B3 B4
(all beams are of length L)
3
δ b = B1P(EI)
L PL
δ s = B2 (AG) Mx = PBL3 Tx = BP4
eq eq
Cantilever,
end load, 3 1 1 1
P
Cantilever,
Uniformly distributed load, 8 2 2 1
q = P/L
Three-point bend,
central load, 48 4 4 2
P
Three-point bend,
384
Uniformly distributed load, 5
8 8 2
q = P/L
Ends built in,
Central load, 192 4 8 2
P
Ends built in,
Uniformly distributed load, 384 8 12 2
q = P/L
(11.24)

In Table 11.24, Mx is the maximun bending moment and Tx is the maximum

shear (transverse) force.

156
 Figure 11.9: Cantilever beam with end point load.

11.7. Example of beam calculations

A cantilever beam has Gc = 30 MPa (80kg/m3 PVC foam), Ef = 200000 MPa
(steel), L = 500mm, tf = 1 mm, tc = 30 mm (d = tc + tf = 31 mm), b = 1 mm,
as in Figure 11.9.
a) Find the total deflection, δ, and the deflection due to shear in percentage of
the total deflection.
The total deflection is described as
P L3 PL
δ = δb + δs = + ,
3 (EI)eq (AG)eq

where the bending contribution is

P L3 P · L3 P · (500)3
δb = = ³ ´ = ³ 2
´ = 0.43P (mm/N)
3 (EI)eq E bt d2
3 f 2f 3 200000·1·1·(31)
2

and the shear contribution is

PL PL P · 500
δs = =³ 2 ´=³ 2
´ = 0.52P (mm/N).
(AG)eq bd Gc 1·(31) ·30
tc 30

The total deflection is

δ = δ b + δ s = 0.43P + 0.52P = 0.95P (mm/N).

157
The deflection due to shear in percentage of the total deflection is
δs 0.52
100 = (100) = 54.7% (mm/N).
δb + δs 0.95

b) Compare a) with a homogenous steel beam with a rectangular cross-section

of height h = 30 mm and width b = 1mm.
The total deflection is
P L3 PL
δ = δb + δs = + ,
3 (EI)eq (AG)eq
where the bending contribution is
P L3 P L3 4P L3 4P · 5003
δb = = ¡ Ebh3 ¢ = = = 0.0926P
3 (EI)eq 3 12 Ebh3 200000 · 1 · 303
and the shear contribution is
PL PL P 500
δs = = ¡ ¢³ ´ = ¡ ¢³ ´ = 0.00026P,
(AG)eq bh E 1·30 200000
k 2(1+ν) 1.2 2(1+0.3)

where k is a shear factor. For rectangular homogeneous cross-sections k is 1.2.

Poissons ratio ν = 0.3. The total deflection
δ = δb + δ s = 0.0926P + 0.00026P = 0.0 929P.
The deflection due to shear in percentage of the total deflection is
δs 0.00026
100 = (100) = 0.28%.
δb + δs 0.0 929
This explains why elementary beam theory usually neglects the contribution
of the transverse shear deformation.

c) Compare a) with a homogenous steel beam with a rectangular cross-section of

height h = 2tf = 2 mm.
Here, the bending contribution is
P L3 P L3 4P L3 4P · 5003
δb = = ¡ Eh3 ¢ = = = 312.5P
3 (EI)eq 3 12 Eh3 200000 · 23

158
and the shear contribution is
PL PL P 500
δs = = ¡ ¢³ ´ = ¡ ¢³ ´ = 0.0039P.
(AG)eq bh E 1·2 200000
k 2(1+ν) 1.2 2(1+0.3)

The total deflection

δ = δ b + δ s = 312.5P + 0.0039P = 312.5039P.

The deflection due to shear in percentage of the total deflection is

δs 0.0039
100 = (100) = 0.00125% .
δb + δs 312.5039

159
 Figure 11.10: A simply supported three-point bended beam.

11.8. Strength and stiffness design example

A simply supported three-point bending beam, with uniform load case, has P =
qL = 10000N, Gc = 40MPa (80kg/m3 PVC foam), (Ec = 100 MPa), shear strength
τ c,cr = 1.5MPa, Ef = 20000MPa (GRP), strength σ f,cr = 100MPa, L = 1000mm,
tc = 50mm, b = 100 mm, see Figure 11.10. The maximum allowed deformation of
the beam is L/50.We assume Ec << Ef and tf << tc .
Find the face thickness, tf .

We know that
Ef btf d2 bd2 Gc
(EI)eq = and (AG)eq = ,
2 tc
and from Table 11.24 we find that
P PL
|Tx |max = and |Mx |max = . (11.25)
2 8

i) Strength design:
Calculate the face thickness, tf , for maximum allowable stress.

160
 We have the formulae
¡PL¢ ¡ 10000·1000 ¢
Mx max tc Ef t c E f 50 · 20000
σf = = ³8 ´= ³8 ´ =
2 (EI)eq 2 f 2f
E bt d2
2
20000·(100)·tf ·50
2

2
¡ 10000·1000 ¢
50 · 20000 250.0
= ³8 ´ = ,
2
20000·(100)·tf ·502 tf
2

here d ≈ tc , thus the face thickness for maximum allowable stress

250 250
tf = = = 2.5 mm.
σf 100
The maximum shear stress in the core is
P
Tx max P 10000
τc = = 2 = = = 1MPa
db db 2db 2 · 50 · (100)
(which is satisfactory since τ c,cr = 1.5 MPa). Next, we want to check the wrinkling
stress 12.1
p √3
σ cr = 0.5 3 Ef Ec Gc = 0.5 20000 · 100 · 40 = 215 MPa

(which is satisfactory since σ cr > σ f = 100 MPa ).

ii) Stiffness design

Find the face thickness for allowable deformation. Total deflection is

δ = δb + δs.

We find that
P L3 PL
δb = 384 and δ s =
5
(EI)eq 8 (AG)eq
such that
P L3 PL P L3 PL
δ = δb + δs = 384 + = ³ ´+ ³ ´=
(EI)eq 8 (AG)eq 384 Ef btf d2 2
5 5 2
8 bdtcGc
10000 · 10003 10000 · 1000 52.1
= ³ 2
´+ ³ 2
´= + 6.25.
384 20000·100·tf ·(50)
8 100·(50) ·40 tf
5 2 50

161
Thus, when allowed deformation δ = L/50
52.08333 3 52.083333
tf = = ¡ 1000 ¢ = 3.8 mm
(δ − 6.25) 50
− 6.25

which is the thickness we must use to fulfill the requirements!

Check the requirements due to the approximations. Since
µ ¶2 µ ¶2
d 53.8
3 = 3 = 601.3 > 100
tf 3.8
d 53.8
(or = = 14.2 > 5.77)
tf 3.8

then the first term of 11.3 is less than 1% of the second, and can therefor be
neglected, and since

6Ef tf d2 6 · 20000 · 3.8 · (53.8)2

= = 105. 6 > 100
Ec t3c 100 · (50)3

then the third term of 11.3 is less than 1% of the second , and can therefor be
neglected such that we can use the reduced form of 11.3. The assumption

Ef btf d2
(EI)eq = D =
2
is valid!

162
12. Failure modes of sandwich panel
A sandwich panel must have the strength to carry the design loads without failing
in one of the possible failure modes. We have to design against and consider all
the failure modes to be sure of that the structure will not fail. Examples on failure
modes is shown in Figure 12.1. A sandwich construction will fail by the failure

Figure 12.1: Some failure modes; a)face yielding/fracture, b) core shear failure, c) and
d) face wrinkling, e) general buckling, f) face dimpling, and g) local indentation.

mode which occurs at the lowest load. The optimum design is when two or more
failure modes occur at the same load. The failure modes can be found on the basis
of when the mode occur. Some of the failure modes is described in the following
chapter.
The skin and core materials should be able to withstand the tensile, compres-
sive and shear stresses induced by the design load. Also the adhesive must be
capable of transferring the shear stresses between skin and core. The sandwich
panel should also have sufficient bending and shear stiffness to prevent excessive
deflection.

163
12.1. Failure loads and stresses
(I) Face Yielding/fracture:
Face yielding/fracture occurs when the normal stress in the face σ f equals or
exceeds the (yield) strength of the face material, σ yf , such that:
Mx PL
σf = = ≥ σ yf .
btf d B3 btf d

(II) Face wrinkling:

Face wrinkling (local buckling) occurs when the normal stress in the face reaches
the wrinkling stress (the local instability stress). Wrinkling occurs when the com-
pressive stress in the face is
Mx PL p
σf ≥ = = 0.5 3 Ef Ec Gc . (12.1)
btf d B3 btf d

Hence, the wrinkling load is independent of the sandwich geometry, and is only a
function of the face and core properties. It is the core that has most influence on
the wrinkling load.

(III) Core shear failure:

Core shear failure occurs in a foam with a plastic-yield point when the principal
stresses satisfy the yield criterion. If the shear stress in the core is large compared
to the normal stress, failure occurs when the shear stress, τ c , equals or exceeds
the yield strength of the foam in shear, τ yc . The core failure is given by
Tx P
τ c max = = ≥ τ yc .
bd B4 bd

(IV) Failure of the adhesive bond (debonding):

Failure of the adhesive bond can occur due to overloading. Debonding (the adhe-
sive between the skin and the core fail) is the most difficult of the mechanisms to
analyze. The adhesive must have a strength equal or bigger than the shear stress
in the bonding line under loading which is almost the same as τ c max . To avoid
debonding therefore
Tx
τ c max = ≤ τ ya
bd

164
where τ ya is the yield shear stress in the adhesive. High thermal stresses, fatigue,
and aging are some of the reasons to debonding.

(V) Core indentation:

Core indentation is only a problem when loads are very localized and can be
avoided if we ensure that the load is distributed over a minimum area of at least
P
A≥ ,
σ yc
where σ yc is the compressive strength of the core.

(VI) General buckling:

General buckling can occur in sandwich constructions due to the transverse shear
deformation. The transverse shear deformation must be accounted for, since this
decreases the buckling load compared with the ordinary Euler buckling cases. The
critical buckling load for sandwich columns with thin faces is given by:
1 1 1
= + ,
Pcr Pb Ps
where Pb is the buckling load in pure bending, and Ps in pure shear, and they are
given as follows
π 2 (EI)eq
Pb = and Ps = (AG)eq ,
(βL)2
where β is the factor depending on the boundary conditions in Euler buckling, see
Figure 12.2

(VII) Face dimpling (local buckling, or intercellular buckling):

Face dimpling may occur in sandwich structures with honeycomb or corrugated
as core material. For a square honeycomb this buckling stress equals
µ ¶2
tf
σ f = 2.5Ef for Poissons ratio ν f = 0.3,
a
where a is the length of the side of the cell. For hexagonal honeycombs the
buckling stress equals
µ ¶2 Ã µ ¶2 !
2Ef tf tf
σf = 2
≤ σ yf , when ν f = 0.3 then σ f ≈ 2. 2Ef ,
1 − νf s s

165
 Figure 12.2: Different cases for Euler buckling.

where s is the radius of the inscribed circle in the honeycomb cell.

(VIII) Fatigue:
Fatigue is said to cause more than 90% of all structural failures. For the face
material, a conservative way to use the fatigue limit under which the material
can undergo an infinite number of load cycles without exhibiting any damage by
taking the allowable face stress σ yf as the material fatigue stress at the given
number of load cycles and stress ratio. For the core material the reasoning is
similar; substitute the allowable shear stress τ yc with the fatigue limit. Be aware
that there is not always data for all materials available. Hopefully, more data will
be available in the future.

166
12.2. Failure-mode maps
Failure-mode maps can be used to design sandwich constructions in a way that
will improve the performance of the sandwich so that no single component is over-
designed. The designer can choose the anticipated failure mode, or making two
different failure modes equally likely occur. Also, this is an advantage for cases
where certain failure modes should be avoided.
The dominant failure mode mechanism for a given design, is the one giving
failure at the lowest load.
A transition in failure mechanism takes place when two or more mechanisms
have the same load. This information can be displayed as a diagram or map
(failure-mode map).The most important transitions we get from equating pairs
of the failure-mode equations are: face yielding - face wrinkling, face yield - core
shear and face wrinkling - core shear, as shown in Figure 12.4. Failure-mode maps
can be constructed from the failure-mode equations that comes out as a result of
the analysis of the different failure modes.
Some different failure modes with the corresponding failure loads for a rectan-
gular sandwich beam are shown in the table below.
Summary of failure modes and failure loads:

Failure Mode Failure Load

B bt d
Face Yielding/fracture (I) P ≥ σ yf 3L f
B bt d p
Face Wrinkling (II) P ≥ 3L f 0.5 3 Ef Ec Gc
Core shear failure (III) P ≥ B4Lbtc
Failure of the adhesive bond (debonding) (IV) Tx ≥ τ ya bd
Core indentation (V) P ≥ σ yc A
(12.2)

12.2.1. Transition equation between face yielding and face wrinkling

Face yielding/fracture occurs when
Mx PL P
σf = = = t = σ yf .
btf d B3 btf d B3 b Lf d
hence the failure load P is given by
µ ¶
tf
P = σ yf B3 bd . (12.3)
L

167
Face wrinkling (local buckling) occurs when

|Mx | PL p
σf = = = 0.5 3 Ef Ec Gc .
btf d B3 btf d

Thus the failure load P can be written as

µ ¶
tf p
P = 0.5B3 bd 3 Ef Ec Gc . (12.4)
L

Putting (12.3) and (12.4) equal to each other, we obtain that

µ ¶ µ ¶
tf tf p
σ yf B3 bd = 0.5B3 bd 3 Ef Ec Gc , (12.5)
L L
i.e 1 1 1 p
σyf = 0.5Ef3 Ec3 Gc3 = 0.5 3 Ef Ec Gc . (12.6)
By using the fact that for most foams and honeycombs the mechanical prop-
erties vary with the density of the material in the following way

Ec = CE ρnc , Gc = CG ρnc , σ yc = Cσ ρm m
c and τ yc = Cτ ρc , (12.7)

where the constants CE , CG , Cσ , Cτ , n and m depend on the type of microstructure

and the mechanical properties of the micromaterial in the core, we find that
p q p
σ yf = 0.5 Ef Ec Gc = 0.5 3 Ef (CE ρnc ) (CG ρnc ) = 0.5 3 Ef CE CG ρ2n
3
c .

Thus the transition between face yielding and face wrinkling is given by the equa-
tion s
³ σ ´3 1
yf
ρc = 2n , (12.8)
0.5 Ef CE CG
which is independent of tf /L and therefore will appear as a horizontal line in a
failure mode map in Figure 12.4, where the variable tf /L is along the horizontal
axis and ρc is along the vertical axis.

168
12.2.2. Transition equation between face yield and core shear
Core shear failure occurs when
Tx P
τ c max = = ≥ τ yc ,
bd B4 bd
where
P = τ yc B4 bd = (Cτ ρm
c ) B4 bd. (12.9)
by the use of (12.7).

Face yielding/fracture occurs when

Mx PL
σf = = = σ yf .
btf d B3 btf d
Thus, we find that the failure load
µ ¶
tf
P = σ yf B3 b d. (12.10)
L

Putting the two expressions in (12.9) and (12.10) equal to each other
µ ¶
m tf
(Cτ ρc ) B4 bd = σyf B3 b d,
L

we obtain that µ ¶
B3 tf
ρm
c = σ yf .
Cτ B4 L
Thus the transition equation between core shear failure and face yielding will be
given by
r µ ¶ m1
m B3 tf
ρc = σ yf . (12.11)
Cτ B4 L

12.2.3. Transition equation between face wrinkling and core shear

Face wrinkling (local buckling) occurs when
Mx PL p
σf = = = 0.5 3 Ef Ec Gc .
btf d B3 btf d

169
Hence,
µ ¶ µ ¶ q
tf p tf
P = 0.5B3 3
bd Ef Ec Gc = 0.5B3 bd 3 Ef (CE ρnc ) (CG ρnc )
L L
µ ¶
tf p
= 0.5B3 bd 3 Ef CE CG ρ2n
c (12.12)
L
by the use of (12.7).
Core shear failure occurs when
Tx P
τ c max = = = τ yc .
bd B4 bd
Hence,
P = τ yc B4 bd = Cτ ρm
c B4 bd (12.13)
by (12.7).
Putting the two expressions in (12.12) and (12.13) equal to each other, we get
that µ ¶
1 tf p m
B3 bd 3 Ef CE CG ρ2n
c = Cτ ρc B4 bd.
2 L
i.e. p µ ¶
B3 3 Ef CE CG tf ¡ 2n ¢− 13 m− 2n
= ρmc ρc = ρc 3 .
2Cτ B4 L
Thus, the transition equation between face wrinkling and core shear failure will
be s p
µ ¶ 1
m− 2n
3
B3 3 Ef CE CG tf m− 2n3
ρc = . (12.14)
2Cτ B4 L

Summing up:
The Face yield - Face wrinkling transition equation is given by
s
³ σ ´3 1
yf
ρc = 2n
. (1) (12.15)
0.5 Ef CE CG

The Face yield - Core shear transition equation is

r µ ¶ m1
m B3 tf
ρc = σ yf . (2) (12.16)
Cτ B4 L

170
 The Face wrinkling - Core shear transition equation is given by
s p
µ ¶ 1
2n
m− 3 B3 3 Ef CE CG tf m− 2n3
ρc = (3) (12.17)
2Cτ B4 L

The transitions of failure modes in (12.15), (12.16) and (12.17) are illustrated
in the failure mode map in Figure 12.4.

An example of a failure mode map To illustrate that the transition between

the failure modes can form a failure mode map, we will have a look at a beam in
three-point bending. The beam has faces with ultimate strength σ yf = 150MPa
and Ef = 70000MPa. B3 = 4 and B4 = 2. We assume a linear relation between
the core properties and the core density, which means that n = m = 1, and
that CE = 1, CG = 0.4 and Cτ = 0.015. Then we obtain the following transition
equations:
The Face yield - Face wrinkling transition equation is given by
s s
³ σ ´3 1 ³ σ ´3 1
yf yf
ρc = 2n
= 2
(1) (12.18)
0.5 Ef CE CG 0.5 Ef CE CG
sµ ¶3
2 150 1
= = 31.053.
0.5 (70000) (1) (0.4)
The Face yield - Core shear transition equation is
µ ¶
(4) tf
ρc = (150) = 20000 . (12.19)
(0.015) (2) L
The Face wrinkling - Core shear transition equation is given by
s p
µ ¶ 1
2n
m− 3 B3 3 Ef CE CG tf m− 2n3
ρc = (12.20)
2Cτ B4 L
s p µ ¶1
3
1 (4) (70000) (1) (0.4) tf 13
= 3

2 (0.015) (2) L
à p !3 µ ¶3
(4) 3 (70000) (1) (0.4) tf
=
2 (0.015) (2) L
µ ¶3
9 tf
= 8.296 4 · 10 .
L

171
 Summing up we obtain that:
The Face yield - Face wrinkling transition equation is given by

log10 ρc = log10 31.053

The Face yield - Core shear transition equation is

µ ¶
tf
log10 ρc = log10 20000 + log10 .
L

The Face wrinkling - Core shear transition equation is given by

µ ¶
¡ 9
¢ tf
log10 ρc = log10 8.296 4 × 10 + 3 log10 .
L
³ ´
t
Thus, substituting x = log10 Lf and y = log10 ρc we obtain that
The Face yield - Face wrinkling transition equation is given by

y = log10 31.053

The Face yield - Core shear transition equation is

y = log10 20000 + x.
The Face wrinkling - Core shear transition equation is given by
¡ ¢
y = log10 8.296 4 × 109 + 3x.

The graphs of these expressions are illustrated in Figure 12.3, where we also have
indicated the failure mechanisms which are pairwise dominating on the respective
side of each graph (1,2 and 3 denote face yield, face wrinkling and core shear,
respectively). For each failure mechanism we now find the region for which
this mechanism is dominating all other mechanisms. This region must be the
intersection of the regions where the failure mechanism is pairwise dominating
the others. The resulting regions are shown in Figure 12.4.

172
 Figure 12.3: The graphs of the three transisition equations.

Figure 12.4: Failure mode map for a sandwich beam in three-point bending, which has
faces made of aluminium and a core with properties that vary with the core density.

173
13. Design Procedures
Designing a sandwich element is very often an integrated process of sizing and
material selection in order to get some sort of optimum design with respect to
the objective you have chosen for instance weight, strength, or stiffness. Note
that all material systems have both advantages and disadvantages. Therefore it is
difficult to state some general terms about choosing materials. But, some material
related properties can still be considered despite the choice of material, such as
density of the core material. An optimum design of a sandwich construction is
very difficult to obtain because there are so many different constraints that the
problem becomes complex. But, considering the most important constraints and
using a simple optimization technique could be very useful in the design process.
An optimization on strength only does not ensure that the sandwich panel is
stiff enough. In most studies done in optimization they do not take into account
all possible failure modes as they should have.

13.1. The stiffness of sandwich structures and its optimization

We will consider a sandwich beam given a load P in three-point bending see Figure
13.1. For several core microstructures it appears that the elastic moduli of the

Figure 13.1: A sandwich beam.

core Ec and Gc vary with the relative density of the core ρc /ρs in the following

174
way µ ¶2 µ ¶2
ρc ρc
Ec = C1 Es , Gc = C2 Es (13.1)
ρs ρs
where ρc is the density of the core, ρs is the density of the solid material (cell
wall) in the core, Es is the Youngs modulus of the solid material (cell wall) in the
core, and C1 (≈ 1) and C2 (≈ 0.4) are proportionality constants. We will explain
the concept of relative density ρc /ρs in more detail later.
Recall that for most cases the flexural rigidity
Ef bt3f Ec bt3c Ef btf d2
(EI)eq = + +
6 12 2
is reduced to
Ef btf d2
(EI)eq = , (13.2)
2
and that the equivalent shear rigidity
bd2 Gc
(AG)eq =
tc
when d ≈ tc , is reduced to
(AG)eq = bdGc . (13.3)
Recall also that the deflection δ
P L3 PL
δ = δb + δs = + . (13.4)
B1 (EI)eq B2 (AG)eq

If we insert (13.2) and (13.3) in (13.4) we obtain that the compliance (the inverse
of the stiffness) of the beam

δ 2L3 L
= 2
+ . (13.5)
P B1 (Ef btf d ) B2 (bdGc )

In most sandwich design the key issue is to minimize the weight, W , (also
called the ”objective function” ) given by

W = mgV = 2ρf gbLtf + ρc gbLtc ≈ 2ρf gbLtf + ρc gbLd, (13.6)

| {z } | {z } | {z } | {z }
facings core facings core

175
for a given bending stiffness P/δ, where m is the mass, V is the volume, g is the
acceleration due to the gravity and ρf and ρc is the density of the face and core
material, respectively. The only free variables are d, tf and ρc . For simplicity we
assume that the core density ρc , is fixed. Then the optimization problem is quiet
easy. By using (13.5), we find that

2L3
tf = ³ ´ (13.7)
B1 Ef bd2 Pδ − L
B2 (bdGc )

and by substituting it into (13.6) we have that the objective function (W is to be

minimized)
⎛ ⎞
3
2L
W = 2ρf gL ⎝ ³ ´ ⎠ + ρc gLd. (13.8)
2 δ L
B1 Ef bd P − B2 (bdGc )
⎛ ⎞
4ρf gL4
W =⎝ ³ ´ ⎠ + ρc gLd
2 δ L
B1 Ef bd P − B2 (bdGc )
We minimize the weight W with respect to the only other free variable, d, by set-
ting ∂W/∂d = 0 and then we obtain the optimum core thickness, dopt . Rearranging
(13.8) we get that
1
W = B1 Ef bδ 2 B1 Ef bL
+ ρc gLd =
4P gρf L4
d − 4gρf L4 B2 (bGc )
d
1
= + ρc gLd,
k1 d2 − k2 d
where
B1 Ef bδ B1 Ef bL
k1 = 4
, k2 = .
4P gρf L 4gρf L4 B2 (bGc )
Differentiating we get
∂W 2k1 d − k2
=− + ρc gL = 0
∂d (k1 d2 − k2 d)2
i.e. ¡ ¢2
− (2k1 d − k2 ) + ρc gL k1 d2 − k2 d = 0,

176
which is a 4th order equation in d.

If the core density also should be treated as a variable the design optimization
problem is more complex.

13.1.1. Example of minimum weight design for given stiffness

This is a method for finding the optimum face and core thickness for a given
stiffness, providing the materials included the core properties are predetermined.
We assume that the core properties can be varied by choosing different densities.

If the compliance δ/P from (13.5) is denoted C, and solving it with respect to
tf as a function of d we obtain from (13.7) that

2L3 2L3
tf = ³ ´= ³ ´
B1 Ef bd2 Pδ − L
B2 (bdGc )
B1 Ef bd2 C − L
B2 bdGc

and by rearranging we get that

⎡ ⎤ ⎡ ⎤
3 3
2L ⎣³ 1 ⎦ = 2L ⎣ ³ 1
tf = ´ ´⎦ =
B1 Ef bd2 L
C − B2 bdG B1 Ef b Cd2 − dL
c B2 bGc
⎡ ⎤
∙ 2 ¸−1
2L2 ⎣ 1 2L 2
Cd d
= ³ ´⎦ = −
B1 Ef b Cd2
− d B1 Ef b L B2 bGc
L B2 bGc

which when substituting into the weight equation, we obtain that the total weight
W of the beam as a function of d (when d = tc )
à ∙ ¸−1 !
¡ ¢ 4ρf L2 Cd2 d
W [g] = 2ρf tf + ρc d L = − + ρc d L. (13.9)
B1 Ef b L B2 bGc

A simply supported beam in three-point bending should have a maximum

mid-point deflection of δ = L/100. B1 is 48 and B2 is 4. Let L = 500mm, the
width b = 1mm the faces are made of aluminium alloy with ρf = 2700kg/m3 and
Ef = 70000MPa and the core has a density ρc = 100kg/m3 and a shear modulus
Gc = 40MPa. Assume a point load of 10N is acting in the middle of the beam.

177
 weight W [g]
2.625

2.5

2.375

2.25

2.125

20 25 30 35 40 45 50

d [mm]

Figure 13.2: Beam weight (g/mm thickness) as function of thickness d.

The weight of the beam per unit thickness (eq. (13.9) times the length L) can
then be plotted versus thickness d as illustrated in Figure 13.2. The weight
⎛ ⎡ ³ 500 ´ ⎤−1 ⎞
2
−3 2
⎜ 4 (2.7 · 10 ) (500) ⎣ 10
100
d d ⎦ ⎟
W (d) [g] = ⎝ − + 0.1 · 10−3 d⎠ 500
48 (70000) 500 4 (40)

simplifying and solving we get that

µ ¶
8.0357 × 10−4
W (d) [g] = + 0.0001d 500.
0.001d2 − 0.00625d
The weight of the beam per unit thickness (equation )
The curve in Figure 13.2 shows that the optimum design is when d ≈ 29mm.
This value can be found more precisely by solving the equation
W 0 (d) = 0
numerically, i.e. solving the equation
1 2.511 2 × 10−3 − 8.035 7 × 10−4 d
+ 0.05 = 0.
d2 (0.001 d − 0.006 25)2

178
The solution of this equation turns out to be i.e. d = 28.617. Thus the minimum
total weight

W (28.617) =

µ ¶
8.0357 × 10−4
= + 0.0001 (28.61729) 500 =
0.001 (28.61729)2 − 0.00625 (28.61729)
= 2.058 6g.
Substituting back gives the face thickness
∙ 2 ¸−1
2L2 Cd d
tf = − =
B1 Ef b L B2 bGc
⎡ ³ 500 ´ 2
⎤−1
2 (500) 2 100
(28.617) 28.617 ⎦
= ⎣ 10 − =
48 (70000) 1 500 4 (40) 1

= 0.23249mm.

179
Part III
Cellular Solids
14. Some definitions of cellular solids
Cellular solids are described by the geometric structure of the cells, that is both
shape and size of the cells and the way the cells are distributed. Foams are three-
dimensional cellular solids and are more complex than the two-dimensional struc-
tures like honeycomb-structures. But, by studying the two-dimensional structures
we make a basis in understanding the more difficult and complex three-dimensional
structures of for instance foams.
One of the most important feature of a cellular solid is what we call the relative
density, defined by
ρ∗
ρ= (14.1)
ρo
where ρ∗ and ρo are the density of the cellular material and the outer (connected
solid) material (which is the material that the cell-walls are made up of), respec-
tively. When the relative density increases, the cell wall gets thicker. Relative
density of the outer material is the same as the volume fraction of the outer
material, po , defined by

volume of outer material (m3 ) Vo

po = 3 = ∗
volume of foam (m ) V

(see Figure 14.1). In order to see this, we observe that the density of the foam

∗ mass of foam (kg) m∗

ρ = = ∗
volume of foam (m3 ) V

and the density of the outer connected material

mass of outer (kg) mo

ρo = 3 = .
volume of outer (m ) Vo

For the case when the inner material in Figure 14.1 and 14.2 is air, we have that

m∗ = mo .

180
 Figure 14.1: Cellular solids.

Figure 14.2: Volume fraction of square honeycomb cells.

Thus, the volume fraction of the outer material

mo m∗
Vo ρo ρo ρ∗
po = ∗ = m∗ = m∗ = . (14.2)
V ρ∗ ρ∗
ρo

Moreover, the volume fraction of the inner material is

volume of inner (m3 ) VI

pI = 3 = .
volume of foam (m ) V∗

Note also that

Vo V ∗ − VI VI
po = ∗
= ∗
= 1 − ∗ = 1 − pI .
V V V
When the cells are square honeycombs, as in Figure 14.2, the volume fractions can
also be expressed in the terms of side length l and wall thickness t in the following
way

181
 Materials ρ
Special ultra-low-density foams 0,001
Polymeric foams (packaging and insulation) 0,05-0,20
Cork 0,14
Softwoods 0,15-0,40
Table 14.1: Relative density for some cellular solids.

µ ¶2 µ µ ¶¶2
(l − t)2 l−t t
po = 1 − pI = 1 − 2
=1− =1− 1− =
l l l
à µ ¶ µ ¶2 ! µ ¶ µ ¶2
t t t t
1− 1−2 + =2 − ,
l l l l
which is approximately 2 (t/l) for small values of t/l. For a closed-cell structure,
the relative density ρ∗ /ρo can be written as
µ ¶
ρ∗ t
= po = 2 .
ρo l

If the relative density ρ∗ /ρo (= po ) is is large, say po > 0.3, the structure will
look more like a solid material with isolated pores than a cellular structure (see
[5], p. 2). Gibson and Ashby refer to materials with relative density less than 0.3
as true cellular solids. Throughout we assume a low density, so that t << l (wall
thickness<<wall-length), see Figure 14.3 and that all the walls have the same
thickness. Simple beam theory is valid only when t/l < 1/4 (see [5], p. 110).
Some examples of the relative density of some cellular solids are given in Table
14.1.

In the same way we define the relative Young’s modulus as

E∗
E=
Eo
where E ∗ and Eo are the Young’s modulus of the cellular material and the solid
material, respectively.

182
 l

h
t

x2

x3

x1

Figure 14.3: Honeycomb structure.

14.1. Mechanics of honeycombs

The cells in a honeycomb-structure are usually hexagonal, but they can also be
triangular or square or rhombic. We will mainly consider the regular square and
regular hexagonal honeycombs. The honeycombs can be made of different types
of materials, such as metal, wood, polymer and ceramics or a combination of two
or more of these materials. For instance, foams of polymer inside the honeycomb
and thin metal on the outside will give a much more stiff/stabile construction so
that the wall thickness of the metal can be made even thinner to obtain an even
more light-weight structure.
When honeycombs are used in load-bearing structures it is important to un-
derstand the mechanical behavior of these two-dimensional structures, illustrated
as hexagonal honeycomb-structures in Figure 14.3.
Besides, many natural three-dimensional cellular solids (for instance wood)
that normally are too complex to be treated can be idealized and analyzed as
honeycomb-structures.

14.2. In-plane deformation properties, uniaxial loading of hexagonal

honeycombs
The study of the in-plane (also called transversal) properties will highlight the
different deformation and failure mechanisms of the cellular solids. In-plane prop-
erties are defined as (the stiffness and strength) properties in the X1 − X2 plane,
see Figure 14.4.

183
 σ2

σ1

σ1

σ2

Figure 14.4: Honeycombstructure loaded in X1 or X2 −direction.

In-plane compression of honeycombs gives us first a bending of the cell walls

and we obtain a linear elastic deformation. If the compression increases beyond
a critical strain the cell walls will undergo collapse by elastic buckling, plastic
yielding, creep or brittle fracture depending on the type material the cell wall is
made of.
If the honeycomb is exposed for tension the cell walls first bend but they will
not obtain elastic buckling. Instead the cell walls will show extensive plasticity
and if the cells are brittle it will fracture. The in-plane stiffness and strength are
the lowest ones because in this plane the cell walls will bend. We will study only
the linear-elastic deformation case of a hexagonal honeycomb in uniaxial loading
in detail.
The study of in-plane properties highlights the mechanisms by which cellular
solids deform and fail.

14.2.1. Linear-elastic deformation

If the hexagonal honeycomb is regular (all angles θ are 30◦ , and wall thicknesses
are equal as Figure 14.5 shows), then the in-plane properties are isotropic. This
means that the in-plane properties of the honeycomb can be described by only
two independent elastic moduli, for instance by Young’s modulus E ∗ and a shear
modulus G∗ . The notation ∗ means the effective value. If the honeycomb is not
regular, the in-plane properties is described with four moduli (e.g. E1,∗ E2,∗ G∗12
and ν ∗12 ). Here, ν ∗12 is the Poisson’s ratio. If the general hexagonal honeycomb
(arbitrary cell wall angle θ) has a low relative density, ρ∗ /ρo , (t/l is small) we have

184
 Figure 14.5: A regular hexagonal honeycomb.

that
t h
ρ∗ ( + 2)
l l
≈ . (14.3)
ρo 2 cos θ( hl + sin θ)
When the honeycomb is regular (h = l and θ = 30◦ ), see Figure 14.5, then (14.3)
is reduced to
t h µ1¶ t
ρ∗ ( + 2)
l l l2 l
(1 + 2) t 2
= h 1 = √ = √
ρo 2 cos θ( l + sin θ) l2 2 23 (1 + 12 ) l 3
which holds when strains are less than 20% or t/l is small.

The response of the honeycomb when loaded in x 1 − or x 2 −direction (see

Figure 14.4), is bending of the cell walls (see Figure 14.6), and is described by five
moduli: two Young’s moduli E1∗ and E2∗ , a shear modulus G∗12 and two Poisson’s
ratio ν ∗12 and ν ∗21 . The reciprocal relation (according to Ashby)

E1∗ ν ∗21 = E2∗ ν ∗12 (14.4)

where ν ∗12 is the negative ratio of the strain in the x 2 − direction to that in the
x 1 −direction for normal loading in the x 1 −direction, reduces the five not inde-
pendent moduli to four independent moduli. (According to Meidell & Lukkassen,
)
ν ∗21 ν∗
= 12 which gives that ν ∗21 E2 = ν ∗12 E1
E1 E2

185
 σ2

σ1 σ1

x2 x2

x1 x1
σ2

Figure 14.6: Honeycomb deformed by normal stresss when loaded in X1 and X2 .

When loading in x 1 − or x 2 − direction, respectively, the four independent

moduli are described as
µ ¶3
E1∗ t cos θ
= ¡h ¢ 2 , (14.5)
Eo l l
+ sin θ sin θ
µ ¶3 ¡ h ¢
E2∗ t l
+ sin θ
= , (14.6)
Eo l cos3 θ
ε2 cos2 θ
ν ∗12 = − = ¡h ¢ (14.7)
ε1 l
+ sin θ sin θ
and
¡h ¢
ε1 + sin θ sin θ
ν ∗21=− = l 2
. (14.8)
ε2 cos θ
For regular hexagonal honeycombs, the two Young’s moduli will be reduced to
µ ¶3
E1∗ E2∗ t
= = 2.3 ,
Eo Eo l
and the two Poisson’s ratio will be reduced to ν ∗12 = ν ∗21 = 1. The honeycomb
exposed to a shear stress is shown in Figure 14.7. It is possible to show that the
shear moduli G∗12 to be
µ ¶3 ¡h ¢
G∗12 t + sin θ
= ¡ h ¢2 l ,
Eo l (1 + 2h/l) cos θ
l

186
 τ τ
x2

x1

Figure 14.7: Honeycomb deformated by shear stress.

which for regular honeycombs is reduced to

µ ¶3
G∗12 t 1 E1∗
= 0.57 =
Eo l 4 Eo
which correctly obeys the relation
E
G=
2 (1 + ν)

for isotropic solids.

14.3. Out-of-plane deformation properties

Out-of-plane properties are defined as (the stiffness and strength) properties in the
x 3 -plane, see Figure 14.8. The function of a honeycomb core in a sandwich panel
is to carry the shear and normal loads in the x 3 -direction. When honeycombs are
loaded either along the x 3 −plane or in out-of-plane shear they are much stiffer
and stronger than if they are loaded in in-plane.
Out-of-plane tension and compression loading of a honeycomb imply that the
cell walls undergo only axial extension or compression and therefore the moduli,
collapse stresses, and strength will be much larger.
Also in the out-of-plane case we will only consider the linear-elastic defor-
mation case, as we did in the in-plane case. The out-of-plane analysis gives the
additional stiffness which is needed for the design of honeycomb cores in sandwich-
panels, and for the description of the behavior of natural honeycomb-like material,
such as wood.

187
Figure 14.8: A honeycomb with out-of-plane loads (faces normal to X3 − direction).

14.3.1. Linear-elastic deformation

A total of nine moduli are needed to describe the out-of-plane deformation, that
means five new ones in addition to the four already described (E1∗ , E2∗ , G∗12 and
υ ∗12 ). We will now find the five new moduli. The Young’s modulus E3∗ , for normal
loading in the x3 -direction is obviously the same as the Young’s modulus for the
outer material, Eo , scaled by the loadbearing section area
E3∗ ρ∗ t
= ≈ .
Eo ρo l
The next two Poisson’s ratio are equal to the solid itself

υ ∗31 = υ ∗32 = υ o ,
and the Poisson’s ratio υ ∗13 and υ ∗23 are found from the reciprocal relations (see
[5], p. 498)
E∗ E∗
υ ∗13 = 1∗ υ o ≈ 0, υ ∗23 = 2∗ υ o ≈ 0.
E3 E3
The shear moduli are more complicated to find because of the non-uniform
deformation in the cell walls due to stress distribution in the honeycomb. The
plane honeycomb may not remain plane, and exact calculations may only be done
using numerical methods. But, we can obtain upper and lower bounds for the two
shear moduli, with the help of the method in [23], by calculating the strain energy

188
associated first with the strain distribution and next by the stress distribution
(see [5], p. 149). If the two coincide, the solution is exact, but if not the true
solution lies somewhere in between the upper and lower bound. We will only give
the results here, for more information see [5], p. 150. The upper and lower bounds
for G∗13 , respectively: µ ¶
G∗13 cos θ t
≤ ,
Go h/l + sin θ l
µ ¶
G∗13 cos θ t
≥
Go h/l + sin θ l
which are identically, and therefor we have an exact solution. For the regular
hexagon we obtain that µ ¶
G∗13 t
= 0.577 . (14.9)
Go l
The upper and lower bounds for G∗23 , respectively:
µ ¶
G∗23 1 h/l + 2 sin2 θ t
≤ ,
Go 2 (h/l + sin θ) cos θ l
µ ¶
G∗23 h/l + sin θ t
≥
Go (1 + 2h/l) cos θ l
which are not identically. They will not coincide for a general, anisotropic honey-
comb but for a regular hexagon they are reduced to
µ ¶
G∗23 t
= 0.577 . (14.10)
Go l

We see that (14.9) and (14.10) are identical, which confirms that regular hexagons
are isotropic in the x 1 −x 2 plane.
The out-of plane shear moduli vary linearly with the relative density (t/l) and
are therefore larger that the in-plane moduli by the factor (t/l)2 .
Now we have all the parameters describing the orthrotropic structure, and we
can put them into the matrix 18.5 in order to find the compliance matrix (the
inverse of the stiffness matrix).

Questions
• What is relative density?

189
• Show that relative density is the same as volume fraction. Assume air inside
the cell.
• Show that if the hexagonal honeycomb has a low relative density, ρ∗ /ρo , (t/l is
small) then
t h
ρ∗ ( + 2)
l l
≈ .
ρs 2 cos θ( hl + sin θ)
• If the relative density of a square honeycomb is 0.8, what is the wall-thickness
of the honeycomb?
• If the volume fraction of a hexagonal honeycomb is 0.8, what is the wall-thickness
of the honeycomb?
• When is simple beam theory valid, according to Ashby & Gibson?
• What is the difference between in-plane and out-of-plane properties of a honey-
comb structure?
• If a hexagonal honeycomb is regular (wall thicknesses are equal), are the in-plane
properties isotropic?
• What does isotropic mean?
• If the honeycomb is irregular and anisotropic, how many moduli for a complete
description of the in-plane properties is needed?
• What is the definition of Poisson’s ratio?

190
Part IV
Mechanics and Effective
Properties of Composite
Structures and Honeycombs
15. On effective Properties of Composite Structures
Strongly non-homogeneous structures have fascinated people for a very long time.
Archaeological observations in Finland show that fibre- reinforced ceramics were
made about 4000 years ago, and that people already at this time had ideas and
theories for intelligent combinations of materials and structures. Analysis of
the macroscopic properties of composites was initiated by the physicists Raleigh,
Maxwell and Einstein. Around 1970 one managed to formulate the physical prob-
lems of material structures and composites in such a way that this field became
interesting from a purely mathematical point of view. This formulation initiated a
new mathematical discipline called homogenization theory. Independently of this
development scientists within the field of micro-mechanics have developed their
own theory concerning the mechanics of composites and structures.
By using these theories we can determine locale and global (effective) proper-
ties of inhomogeneous structures which are too complex to be treated by conven-
tional computational methods. The theories make it possible to design material
structures with optimal properties with respect to weight, strength, stiffness, heat
conductivity, electric conductivity, magnetic permeability, viscosity etc.
Almost all literature on this field assumes relatively high knowledge in math-
ematics. However, in this paper we give a short introduction to the theory which
require physical intuition rather than deep theoretical understanding. We hope
this treatment makes it easier to understand some basic aspects of the theory for
a broader audience, especially people with background in engineering.

16. The thermal problem

The heat flow in the direction of decreasing temperature. Moreover, for isotropic
materiels, the velocity v of the heat flow is proportional to the gradient of the

191
temperature, i.e.
⎡ ∂u ⎤
∙ ¸T λ ∂x1
∂u ∂u ∂u
v = −λ grad u (= − λ ,λ ,λ ∂u ⎦
= − ⎣ λ ∂x ),
∂x1 ∂x2 ∂x2 ∂u
2
λ ∂x2

where u is the temperature and λ is called the conductivity. Assuming no heat

source in the body and no dependence of time, a simple application of the diver-
gence theorem (Gauss theorem) shows that
∂v1 ∂v2 ∂v3
div v = 0, (recall that div v = + + ).
∂x1 ∂x2 ∂x3
Some materials have different effective conductivities in each direction. This
means that the velocity v of the heat flow can be written as
∙ ¸T
∂u ∂u ∂u
v = − λ11 , λ22 , λ33 ,
∂x1 ∂x2 ∂x2
or even more generally as
v = −A grad u,
where A a symmetric matrix of the form
⎡ ⎤
λ11 λ12 λ13
A = ⎣ λ12 λ22 λ23 ⎦ .
λ13 λ23 λ33

In a composite material we distinguish between the local conductivity A (which

varies locally) and the effective conductivity
⎡ ∗ ⎤
λ11 λ∗12 λ∗13
A∗ = ⎣ λ∗12 λ∗22 λ∗23 ⎦
λ∗13 λ∗23 λ∗33

(which is constant). This matrix is found as follows. Imagine that the composite
is subjected to a homogeneous temperature-field, that is a temperature-field such
that the average value hgrad ui of grad u is constant. Letting hvi be the average
value of v = −A grad u we can obtain A∗ from the relation

hvi = A∗ hgrad ui .

192
The average is taken over some representative domain Y of the composite. In
general this volume should be much larger than the typical length-scale of the
material (e.g. fiber-diameter), but in the case of periodic materials we can as
well let Y be the cell of periodicity. Roughly speaking the effective conductivity
matrix A∗ is the matrix of a corresponding homogeneous material with the same
”effective” thermal properties as the actual composite. In particular, the ”the
thermal energy” in Y of the actual composite is equal to ”the thermal energy” in
the corresponding homogeneous material, i.e.
Z
1 1
(grad u · A grad u) dv = hgrad ui · A∗ hgrad ui |Y |. (16.1)
Y 2 2
| {z } | {z }
thermal energy in the composite thermal energy in corresponding
homogenized material.

For example, if we want to find λ∗11 we impose the thermal field such that hgrad ui =
[1, 0, 0]T then measure the energy W inside Y and obtain λ∗11 from (16.1):

W
λ∗11 = 2 .
|Y |

17. Isotropic elastic materials

For an isotropic material the shear modulus G and bulk modulus K in plane
elasticity (plane strain) are related to the well known Young’s modulus E and
Poisson’s ratio ν as follows:
E
K= ,
2 (1 + ν) (1 − 2ν)
E
G= .
2 (1 + ν)
The bulk modulus k of the three-dimensional theory is given by
E 1
k= .
3 1 − 2ν
Thus the plane strain bulk modulus K can be expressed as
G
K =k+ .
3

193
If the material is a thin plate, we consider the plane-stress problem. In this case
the plane stress bulk modulus is expressed as
µ ¶
E 1
K= .
2 1−ν
The shear modulus G is independent of the dimension and also independent of
whether we are dealing with the plane-strain -or plane stress problem.

18. Orthotropic composites

Consider a linear elastic composite material in R3 which is locally orthotropic,
and globally orthotropic. This means that in each point of the structure we have
that the stress-strain relation is given as
⎡ ⎤ ⎡ ⎤⎡ ⎤
σ 11 C1111 C1122 C1133 0 0 0 e11
⎢ σ 22 ⎥ ⎢ C2211 C2222 C2233 0 0 0 ⎥ ⎢ ⎥
⎢ ⎥ ⎢ ⎥⎢ e22 ⎥
⎢ σ 33 ⎥ ⎢ C3311 C3322 C3333 0 0 0 ⎥ ⎢ ⎥
⎢ ⎥=⎢ ⎥⎢ e33 ⎥
⎢ σ 12 ⎥ ⎢ 0 0 0 C1212 0 0 ⎥ ⎢ ⎥
⎢ ⎥ ⎢ ⎥⎢ γ 12 ⎥
⎣ σ 23 ⎦ ⎣ 0 0 0 0 C2323 0 ⎦⎣ γ 23 ⎦
σ 13 0 0 0 0 0 C1313 γ 13
| {z } | {z }| {z }
σ C e

and the average stress-strain relation of the form

⎡ ⎤ ⎡ ∗ ∗ ∗
⎤⎡ ⎤
hσ 11 i C1111 C1122 C1133 0 0 0 he11 i
⎢ hσ 22 i ⎥ ⎢ ∗ ∗ ∗
C2211 C2222 C2233 0 0 0 ⎥⎢ he22 i ⎥
⎢ ⎥ ⎢ ⎥⎢ ⎥
⎢ hσ 33 i ⎥ ⎢ ∗ ∗ ∗
C3311 C3322 C3333 0 0 0 ⎥⎢ he33 i ⎥
⎢ ⎥=⎢ ⎥⎢ ⎥, (18.1)
⎢ hσ 12 i ⎥ ⎢ 0 0 0 ∗
C1212 0 0 ⎥⎢ hγ 12 i ⎥
⎢ ⎥ ⎢ ⎥⎢ ⎥
⎣ hσ 23 i ⎦ ⎣ 0 0 0 0 ∗
C2323 0 ⎦⎣ hγ 23 i ⎦
∗
hσ 13 i 0 0 0 0 0 C1313 hγ 13 i
| {z } | {z }| {z }
hσi C∗ hei

i.e.
σ = Ce, hσi = C ∗ hei . (18.2)

194
Here we assume equilibrium of stresses in each direction:
∂σ 11 ∂σ 12 ∂σ 13
+ + = 0,
∂y1 ∂y2 ∂y3
∂σ 21 ∂σ 22 ∂σ 23
+ + = 0,
∂y1 ∂y2 ∂y3
∂σ 31 ∂σ 32 ∂σ 33
+ + = 0.
∂y1 ∂y2 ∂y3

where σij = σ ji (obtained by applying Gauss theorem as for the heat problem)
These equations follows if we assume that the net force is 0. The symbol h·i
denotes the average taken over some representative domain Y of the composite.
In general this volume should be much larger than the typical length-scale of the
material (e.g. fiber-diameter), but in the case of periodic materials we can as well
let Y be the cell of periodicity. Roughly speaking the effective stiffness matrix C ∗
is the matrix of a corresponding homogeneous material with the same ”effective”
properties as the actual composite. In particular, the actual strain energy stored in
Y is equal to the strain energy stored in the corresponding homogeneous material,
i.e. Z
1 1
(e · Ce) dv = hei · C ∗ hei |Y | . (18.3)
2 |2 {z }
|Y {z }
strain energy in the composite strain energy in corresponding
homogenized material.

∗
Example 18.1. If we want to find C1111 we stretch the periodic structure in such
a way that average strain hei = [1, 0, 0, 0, 0, 0]T , then measure (or compute) the
average stress hσi = [hσ 11 i , hσ 22 i , hσ 33 i , hσ 12 i , hσ 23 i , hσ 13 i]T and finally compute
∗ ∗
the coefficient C1111 from (18.1) which gives us that C1111 = hσ 11 i . Equivalently,
we can ask the FE-program to compute the strain energy W (the left side of
∗
(18.3)) and then compute C1111 from (18.3), i.e. we obtain
Z
1 1 1 ∗
(e · Ce) dv = [1, 0, 0, 0, 0, 0]T · C ∗ [1, 0, 0, 0, 0, 0]T |Y | = C1111 |Y | ,
Y 2 2 2
| {z }
strain energy in the composite

thus
∗ 2(strain energy)
C1111 = .
|Y |

195
 The stiffness matrix is symmetric and the elements may be expressed as follows
(see e.g. [32]):

∗ 1 − ν ∗23 ν ∗32 ∗ ν ∗21 + ν ∗31 ν ∗23 ∗ ν ∗31 + ν ∗21 ν ∗32

C1111 = , C 1122 = , C1133 =
∆E2∗ E3∗ ∆E2∗ E3∗ ∆E2∗ E3∗
∗ ∗ ∗ ∗ ∗
∗ 1 − ν 13 ν 31 ∗ ν 32 + ν 12 ν 31 ∗ 1 − ν ∗12 ν ∗21
C2222 = , C 2233 = , C3333 = (18.4)
∆E1∗ E3∗ ∆E1∗ E3∗ ∆E1∗ E2∗
∗
C1212 = G∗12 , C2323
∗
= G∗23 , C1313∗
= G∗13 ,

where
1 − ν ∗12 ν ∗21 − ν ∗23 ν ∗32 − ν ∗31 ν ∗13 − 2ν ∗21 ν ∗32 ν ∗13
∆= .
E1∗ E2∗ E3∗
Here, ’Ei∗ ’ are the effective Young’s moduli, ’G∗ij ’ are the effective shear moduli
and ’ν ∗ij ’ are the effective Poisson’s ratios. The inverse of the effective stiffness
matrix (the compliance matrix) which (certainly) also is symmetric is given by
⎡ ⎤
1 ν ∗12 ν ∗13
∗ − ∗ − ∗ 0 0 0
⎢ E1ν ∗21 1 E2 E3
ν ∗23 ⎥
⎢ − ∗ − 0 0 0 ⎥
⎢ E∗1 E2 ∗ ∗ E3 ∗
⎥
⎢ ν 31 ν ⎥
⎢ − E ∗ − E32∗ E1∗ 0 0 0 ⎥
⎢ 1 2 3 ⎥. (18.5)
⎢ 0 0 0 1
0 0 ⎥
⎢ G ∗
12 ⎥
⎢ 0 0 0 0 1
0 ⎥
⎣ ∗
G23 ⎦
1
0 0 0 0 0 G∗ 13

19. Square symmetric unidirectional two-phase structure

In the case of square honeycombs with locally isotropic material properties (local
shear moduli Go and GI and local plane strain bulk moduli Ko and KI with
corresponding volume fractions po and pI , respectively, where the subscript o and
I denote outer and inner material, respectively (see Figure 19.1), the stiffness
matrix reduces to a matrix of the form:
⎡ ∗ ⎤
K + G∗T K ∗ − G∗T l∗ 0 0 0
⎢ K ∗ − G∗T K ∗ + G∗T l∗ 0 0 0 ⎥
⎢ ∗ ⎥
⎢ l l ∗
n ∗
0 0 0 ⎥
⎢ ⎥. (19.1)
⎢ 0 0 ∗
0 GT,45 0 0 ⎥
⎢ ⎥
⎣ 0 0 0 0 G∗L 0 ⎦
0 0 0 0 0 G∗L

196
Figure 19.1: The structure of square honeybombs with locally isotropic material prop-
erties.

Figure 19.2: The 4 moduli measure resistence against the indicated average strains.

Here, K ∗ is the effective transverse (also called ”in-plane”) bulk modulus, G∗T ,
G∗T,45 are the effective transverse shear moduli and G∗L is the longitudinal (also
called ”out-of plane”) shear modulus, see Figure 19.2. In this case the compliance
matrix reduces to

197
 ⎡ ν∗ ν∗
⎤
1
ET∗
− ET∗ − EL∗ 0 0 0
⎢ ν∗
T
ν∗
L
⎥
⎢ − ET∗ 1
− EL∗ 0 0 0 ⎥
⎢ T ET∗ L ⎥
⎢ ν∗ ν∗ 1 ⎥
⎢ − EL∗ − EL∗ EL∗ 0 0 0 ⎥
⎢ L L ⎥. (19.2)
⎢ 0 0 0 1
0 0 ⎥
⎢ G∗T,45 ⎥
⎢ 0 0 0 0 1
0 ⎥
⎣ G∗L ⎦
1
0 0 0 0 0 G∗L

Using (18.4) and the symmetry we obtain the relations

ET∗
G∗T = (19.3)
2 (1 + ν ∗T )

4 1 1 4 (ν ∗L )2
= ∗ + ∗+ . (19.4)
ET∗ GT K EL∗
l∗ = ν ∗L 2K ∗ , n∗ = EL∗ + 4 (ν ∗L )2 K ∗ (19.5)
Moreover, it has been proved by Hill [11] that

4(ν o − ν I )2 1 1 1
EL∗ = po Eo + pI EI + ³ ´2 (po + pI − ∗ ), (19.6)
1 Ko KI K
Ko
− K1I

νo − νI 1 1 1
ν ∗L = po ν o + pI ν I − 1 1 (po + pI − ∗ ). (19.7)
Ko
− KI Ko KI K
These two formulae were proved in [11] for the case of transverse isotropy. How-
ever, by following the proof in [11] it is easy to check that the same facts hold in
our case.

19.1. Calculation of stiffness and compliance matrix

We remark that (19.3), (19.4), (19.5), (19.6) and (19.7) hold for all two-component
unidirectional fiber composite (orientated in the direction x3 ) satisfying the prop-
erty of square symmetry, i.e. the case when the stiffness matrix is of the form
(19.1) (we say that the composite satisfies the property of transverse isotropy if
G∗T = G∗T,45 , see Figure 19.3). In order to compute all components stiffness ma-
trix (19.2) we only have to compute (e.g. numerically by using the finite element
method) the four components G∗L , G∗T , G∗T,45 , and K ∗ . The other two moduli

198
Figure 19.3: Hexagonal honeycomb. Exampel of structure satisfying the property of
transverse isotropy (i.e G∗T = G∗T,45 ).

l∗ and n∗ are the found by inserting the values of EL∗ (19.6) and ν ∗L (19.7) into
(19.5). If we also want to know ET∗ and ν ∗T we can evaluate these values by first
evaluating ET∗ from (19.4) and finally ν ∗T from (19.3).
It is possible to show that G∗L can be found exactly as the effective conductivity
in the similar 2-dimensional problem by letting Go and GI play the same role as
the local conductivity for that problem.

Exercise 19.1. Consider square symmetric unidirectional two-phase composite

with volume-fractions pI = po = 1/2, Poissons ratios ν I = ν o = 0.3 and Youngs
modulus EI = 0.5, Eo = 1.(no units)

1. Find KI , Ko , GI , Go .

2. By performing a FE-calculation it is possible to find the following effective

moduli:

K ∗ = 0.658455,
G∗T = 0.273775,
G∗T,45 = 0.259418,
G∗L = 0.274426.

Find the effective stiffness matrix C ∗

3. Find also ν ∗L , EL∗ , ET∗ and ν ∗T . and find the effective compliance matrix.

199
20. Numerical methods for periodic structures
In order to use conventional software to solve displacement field numerically for
periodic structures, e.g. by the finite element method, it is often necessary to
”translate” the information of the average strain hei into equivalent boundary
conditions for the on the cell of periodicity

Y = h0, y1 i h0, y2 i h0, y3 i

for the displacement ⎡ ⎤

u1
u = ⎣ u2 ⎦ .
u3
This is obtained by letting each pair of points (xl− , xl+ ) (the latter point with
the largest coordinates) on opposite faces with normal vector nl be coupled to
each other in such a way that
⎡ ⎤ ⎡ ⎤ ⎡ ⎤
u1 (xl+ ) u1 (xl− ) c1l
⎣ u2 (xl+ ) ⎦ = ⎣ u2 (xl− ) ⎦ + ⎣ c2l ⎦ , (20.1)
u3 (xl+ ) u3 (xl− ) c3l
where ⎡ ⎤
c1l
⎣ c2l ⎦
c3l
is a constant vector. By integrating along the normal vector nl we obtain that
¿ À
∂uk uk (xl+ ) − uk (xl− ) ckl
= = ,
∂xl yl yl
thus, recalling that
∂uk
ekk =
∂xk
∂uk ∂ul
γ kl = +
∂xl ∂xk
we obtain the following useful relation
ckk
= hekk i
yk

200
 Figure 20.1: The Y -cell in the symmetric case.

ckl clk
+ = hγ kl i . (20.2)
yl yk
Hence
ckk = yk hekk i .
Observe that ckl and clk are not uniquely determined by this relation when k 6= l.
Thus in this case these constants can be chosen independently (as long as (20.2)
is satisfied). E.g. we can choose
hγ kl i
ckl = clk = 1 . (20.3)
yl
+ y1k

If the material properties are symmetric in each coordinate with respect to

the midpoint ((1/2)y1 , ..., (1/2)yn ) of the Y -cell problem (see Figure 20.1) then
we can often use simpler boundary conditions than (20.1).
For example, in the case when all average shear strains hγ kl i = 0, we can in
(20.1) use the Dirichlet condition

ul (xl− ) = 0, ul (xl+ ) = cll (= yl hell i), l = 1, 2, 3, (20.4)

and drop all the other boundary conditions. The latter is equivalent with setting
a Neumann condition, ∂ui /∂xl = 0, i 6= l on the same faces. It is easy to
see physically that these simplified boundary conditions hold by considering the
deformation of the whole periodic structure, since it is obvious that the solution
(which indeed represents the deformed body) must inherit the same symmetry
as the material itself (see Figure 20.2). Moreover, in the case when all average
normal strains hekk i = 0, we can in (20.1) use the Dirichlet condition
hγ il i
ui (xl− ) = 0, ui (xl+ ) = cil (= 1 ), i 6= l (20.5)
yl
+ y1i

201
 Figure 20.2: Deformation of the periodic structure in the symmetric case.

and drop all the other boundary conditions. The latter is equivalent with setting
a Neumann condition, ∂ui /∂xi = 0, i 6= l for l = 1, 2, 3 on the same faces.
The boundary conditions for the conductivity case can be simplified in the
same way.

20.1. Coordinate transformation

For the computation of effective moduli it is sometimes convenient to rotate the
original coordinate system. Consider an orthonormal coordinate system with basis
vectors

n1 = [n11 , n12 , n13 ]

n2 = [n21 , n22 , n23 ]
n3 = [n31 , n32 , n33 ] .

A vector with coordinates x = (x1 , x2 , x3 ) (relative to the usual coordinate system)

will have coordinates x0 = (x01 , x02 , x03 ) (relative to the new coordinate system) given
by the relation ⎡ 0 ⎤ ⎡ ⎤⎡ ⎤
x1 n11 n12 n13 x1
⎣ x02 ⎦ = ⎣ n21 n22 n23 ⎦ ⎣ x2 ⎦
x03 n31 n32 n33 x3
(see Figure 20.3). It is possible to show that the following relation between the
£ ¤T
strain e = e11 , e22 , e33, γ 12 , γ 23, γ 13 , γ ij = 2eij (in the usual coordinate system)
£ ¤T
and e0 = e011 , e022 , e033, γ 012 , γ 023 , γ 013 , γ 0ij = 2e0ij (in the new coordinate system):

e0 = Te,

202
 Figure 20.3: The two coordinate systems.

where

⎡ ⎤
n211 n212 n213 n11 n12 n12 n13 n11 n13
⎢ 2
n21 2
n22 2
n23 n21 n22 n22 n23 n21 n23 ⎥
⎢ ⎥
⎢ 2
n31 2
n32 2
n33 n31 n32 n32 n33 n31 n33 ⎥
T=⎢
⎢
⎥.
⎥
⎢ 2n11 n21 2n12 n22 2n13 n23 n11 n22 + n21 n12 n12 n23 + n22 n13 n11 n23 + n21 n13 ⎥
⎣ 2n21 n31 2n22 n32 2n23 n33 n21 n32 + n31 n22 n22 n33 + n32 n23 n21 n33 + n31 n23 ⎦
2n11 n31 2n12 n32 2n13 n33 n11 n32 + n31 n12 n12 n33 + n32 n13 n11 n33 + n31 n13

Moreover, we can obtain a similar relation between the corresponding stresses σ

and σ 0 :
σ 0 = T−T σ,
where T−T is obtained from T by changing the factors of 2 in T symmetrically
about the diagonal, i.e.
⎡ ⎤
n211 n212 n213 2n11 n12 2n12 n13 2n11 n13
⎢ n221 2
n22 2
n23 2n21 n22 2n22 n23 2n21 n23 ⎥
⎢ ⎥
⎢ n 2
n 2
n 2
2n n 2n n 2n31 n33 ⎥
T−T = ⎢ 31 32 33 31 32 32 33
⎢ n11 n21 n12 n22 n13 n23 n11 n22 + n21 n12 n12 n23 + n22 n13 n11 n23 + n21 n13
⎥.
⎥
⎢ ⎥
⎣ n21 n31 n22 n32 n23 n33 n21 n32 + n31 n22 n22 n33 + n32 n23 n21 n33 + n31 n23 ⎦
n11 n31 n12 n32 n13 n33 n11 n32 + n31 n12 n12 n33 + n32 n13 n11 n33 + n31 n13

The stress-strain relation in the new coordinate system can therefore be written
as:
σ0 = C 0 e0 ,

203
where
C = TT C 0 T.
Concerning these facts we refer to e.g. [3, p. 212].
In the plane strain case we put all strains related to the x3 -variable equal to
0. If we also assume that the new coordinate system is obtained from the stan-
dard one by a rotation in the x1 -x2 -plane then we obtain the following simplified
relations:
⎡ ⎤
C1111 C1122 C1112
C = ⎣ C2211 C2222 C2212 ⎦ ,
C1211 C1222 C1212
⎡ ⎤
n211 n212 n11 n12
T = ⎣ n221 n222 n21 n22 ⎦,
2n11 n21 2n12 n22 n11 n22 + n21 n12
⎡ ⎤
n211 n212 2n11 n12
T−T = ⎣ n221 n222 2n21 n22 ⎦.
n11 n21 n12 n22 n11 n22 + n21 n12
As an example, let us consider the square symmetric case, i.e. when the stiffness

matrix is of the form

⎡ ⎤
K ∗ + G∗T K ∗ − G∗T 0
C ∗ = ⎣ K ∗ − G∗T K ∗ + G∗T 0 ⎦
∗
0 0 GT,45
h√ √ i
(c.f. (19.1)). Performing a rotation of 45◦ , i.e. [n11 , n12 ] = 22 , 22 and [n21 , n22 ] =
h √ √ i
− 22 , 22 , the corresponding stiffness matrix in the rotated coordinate system
becomes
⎡ ∗ ⎤
K + G∗T,45 K ∗ − G∗T,45 0
C ∗0 = T−T C ∗ T−1 = ⎣ K ∗ − G∗T,45 K ∗ + G∗T,45 0 ⎦ , (20.6)
0 0 G∗T

i.e. C ∗0 is the same as C ∗ except that the shear moduli G∗T,45 and G∗T have changed
place. This explains the use of the index ”45” and shows that we can calculate
G∗T,45 exactly as we calculate G∗T except by rotating the coordinate system 45◦ .

204
Exercise 20.1. Verify (20.6) by calculating C ∗ = TT C ∗0 T

Exercise 20.2. Let

⎡ ⎤
K ∗ + G∗ K ∗ − G∗ 0
C ∗0 = ⎣ K ∗ − G∗ K ∗ + G∗ 0 ⎦
0 0 G∗

By matrix-multiplication it is possible to obtain that

TT C ∗0 T =C ∗0 . (20.7)

for any orthonormal coordinate system [n11 , n12 ] = [cos φ, sin φ] , [n21 , n22 ] =
[− sin φ, cos φ] This is done by simplifying each of the elements of the matrix.
For example we find that
¡ T ∗0 ¢
T C T 11 = n411 K ∗ + n411 G∗ + 2n211 n221 K ∗ + 2n211 n221 G∗ + n421 k + n421 G∗

Verify that (20.7) holds for this element. (Note that (20.7) explains why compos-
ites with this type of effective stiffness matrixes are called transversely isotropic).

Solution:

¡ T ∗0 ¢ ¡ ¢ ¡ ¢2
T C T 11 = (K ∗ + G∗ ) n411 + 2n211 n221 + n421 = (K ∗ + G∗ ) n211 + n221 =
¡ ¢2
= (K ∗ + G∗ ) cos2 φ + sin2 φ = (K ∗ + G∗ ) 12 = (K ∗ + G∗ ) .

Exercise 20.3. Consider the same square symmetric unidirectional two-phase

composite in Exercise 19.1. Suppose that the composite is used as core material
in a sandwich construction. How should we direct the fibres in order to maximize
the stiffness of the sandwich (compare the three alternatives in Figure 20.4).

Solution: The deflection of the sandwich due to shear deformation in the

core is dependent of G∗L (in alternative 1), G∗T (in alternative 2) and G∗T,45 (in
alternative 3), where these parameters correspond ti the same coordinate system as
in 19.1). Using the values we see that alternative 1 is (insignificantly) better than
alternative 2 which is better than alternative 3. (For example when calculating
the deflection of a sandwich beam of type alternative 1 we use the same formulae
as in the isotropic case by using G = G∗L as the shear modulus of the core and the
same goes for alt. 2 and alt. 3).

205
 Figure 20.4: Sandwich plate with three alternative cores.

21. A computational example

In the case of square honeycombs with locally isotropic material properties (see
Figure 19.1) the structure is symmetric with respect to the midpoint of the Y -cell
Y = [−1, 1]3 in all the coordinates x1 , x2 , x3 , and also in the coordinates x01 , x02 , x3
in the coordinate system obtained by a rotation of 45◦ in the x1 -x2 -plane. The
effective stress/strain-relation (18.1) reduces in this case to
⎡ ⎤ ⎡ ∗ ⎤⎡ ⎤
hσ 11 i K + G∗T K ∗ − G∗T l∗ 0 0 0 he11 i
⎢ hσ 22 i ⎥ ⎢ K ∗ − G∗T K ∗ + G∗T l∗ 0 0 0 ⎥ ⎢ he22 i ⎥
⎢ ⎥ ⎢ ∗ ⎥ ⎢ ⎥
⎢ hσ 33 i ⎥ ⎢ l l ∗
n ∗
0 0 0 ⎥ ⎢ he33 i ⎥
⎢ ⎥ ⎢ ⎥⎢ ⎥
⎢ hσ 12 i ⎥ = ⎢ 0 0 0 G∗T,45 0 0 ⎥ ⎢ hγ 12 i ⎥ . (21.1)
⎢ ⎥ ⎢ ⎥ ⎢ ⎥
⎣ hσ 23 i ⎦ ⎣ 0 0 0 0 G∗L 0 ⎦ ⎣ hγ 23 i ⎦
hσ 13 i 0 0 0 0 0 G∗L hγ 13 i

In order to compute the effective in-plane moduli K ∗ and G∗T we solve the cell
problem for hei = [1, 1, 0, 0, 0, 0, ]T and hei = [1, −1, 0, 0, 0, 0, ]T and compute the
corresponding values K ∗ = hσ11 i /2 and G∗T = hσ11 i /2, respectively. Alternatively
we can compute the strain energy W and compute K ∗ and G∗T from the identity
1
W = hei · C ∗ hei |Y | (21.2)
2

206
(discussed in (18.3)). For the computation it is enough to solve the two-dimensional
cell problem using plane strain. This is due to the fact that the solution of the
three dimensional cell problem in both cases will be independent of the x3 -variable
and we do not use any information of the strain or stresses in x3 -direction for the
computation of K ∗ and G∗T . Because of the symmetries it suffices to solve the prob-
lem on 1/4 of the Y -cell, e.g. on the square Y = [0, 1]2 . Since |Y | = 1, we obtain
from (21.2) that K ∗ = W/2 and G∗T = W/2. We can use uniform boundary con-
ditions on each face of this square (see Section 20), i.e. for hei = [1, 1, 0, 0, 0, 0, ]T
the boundary conditions are :

u1 (0, x2 ) = u2 (x1 , 0) = 0,
u1 (1, x2 ) = u2 (x1 , 1) = 1

and for hei = [1, −1, 0, 0, 0, 0, ]T :

u1 (0, x2 ) = u2 (x1 , 0) = 0,
u1 (1, x2 ) = 1
u2 (x1 , 1) = −1

(this follows from (20.4)). Due to symmetries the module G∗T,45 can be found
exactly as G∗T except that we must rotate the coordinate system 45◦ (see the last
part of Section 20.1).
Alternatively we can compute G∗T,45 more directly by solving the cell problem
for hei = [0, 0, 0, 1, 0, 0]T and computing the corresponding value G∗T,45 = hσ12 i · 1,
or equivalently, compute the strain energy W and find G∗T,45 from the identity
(21.2) (with |Y | = 1), which gives G∗T,45 = 2W . In this case we use the following
boundary conditions on each face of the square ( according to (20.5))

u2 (0, x2 ) = u1 (x1 , 0) = 0,
1
u2 (1, x2 ) = u1 (x1 , 1) = .
2

The module G∗L can be found by using hei = [0, 0, 0, 0, 0, 1]T and computing
the corresponding value G∗L = hσ 13 i . The problem with doing this is that it often
requires a full 3D FE-computation regardless of the fact that the solution of the
cell problem is independent of the x3 -variable. Therefore it is often easier to
solve the corresponding 2D heat-conductivity problem (see Subsection 19.1) with

207
boundary conditions

u(0, x2 ) = 0
u(1, x2 ) = 1

using Go and GI as ”thermal conductivities” and computing G∗L from the thermal
energy W (16.1), which gives
G∗L = 2W.
Once we have computed the 4 moduli K ∗ , G∗T , G∗T,45 , and G∗L we can easily
obtain all the other moduli (see Subsection 19.1).

As an example, consider a square honeycomb structure (see Figure 19.1) with

volume fractions pI = po = 1/2 and locally isotropic material properties ν I =
ν o = 0.3, EI = 0.5, Eo = 1 (or equivalently KI = 0. 48076923, Ko = 0. 96153846,
GI = 0. 19230769, Go = 0.38461538). By performing a FE-calculation according
to the above method we obtain the following effective moduli:

K ∗ = 0.658455,
G∗T = 0.273775,
G∗T,45 = 0.259418,
G∗L = 0.274426.

Hence, by using (19.3), (19.4), (19.6) and (19.7) we obtain that

ν ∗L = 0.3,
EL∗ = 0.75,
ET∗ = 0.70779659,
ν ∗T = 0.29266111,

and by (19.5)

l∗ = 0. 39507,
n∗ = 0.98704.

Thus the effective stiffness matrix is written as

208
 ⎡ ⎤
0.932226 0.384677 0.39507 0 0 0
⎢ 0.384677 0.932226 0.39507 0 0 0 ⎥
⎢ ⎥
⎢ 0.395072 0.395072 0.98704 0 0 0 ⎥
⎢ ⎥.
⎢ 0 0 0 0.259418 0 0 ⎥
⎢ ⎥
⎣ 0 0 0 0 0.274426 0 ⎦
0 0 0 0 0 0.274426
and the effective compliance matrix (19.2) takes the form

⎡ ⎤
1.41284 −0.413482 −0.4 0 0 0
⎢ −0.413482 1.41284 −0.4 0 0 0 ⎥
⎢ ⎥
⎢ −0.4 −0.4 1.33333 0 0 0 ⎥
⎢ ⎥
⎢ 0 0 0 3.8548 0 0 ⎥.
⎢ ⎥
⎣ 0 0 0 0 3.64397 0 ⎦
0 0 0 0 0 3.64397
In this example we used the same Poisson’s ratios in both phases. Let us consider
the case when ν I = 0.4, ν o = 0, EI = 0.5, Eo = 1 or equivalently KI = 0. 89285714,
Ko = 0.5, GI = 0.17857143, Go = 0.5. In this case we obtain the following effective
moduli:
K ∗ = 0.66335 ν ∗L = 0.22386372
G∗T = 0.3038805 EL∗ = 0.79338859
.
G∗T,45 = 0.269322 ET∗ = 0.79193337
G∗L = 0.307502 ν ∗T = 0.3030342
Remark 1. In the first example (where the Poisson’s ratios are the same in both
phases) ν ∗L and EL∗ equals the arithmetic mean of the corresponding phase prop-
erties (usually referred to as ”the law of mixtures”). From the last example we
observe that this may not be true when the Poisson’s ratios of the phases are
different.
Exercise 21.1. Verify the above values for K ∗ , G∗T , G∗T,45 , and G∗L by performing
a FE-calculation according to the above method. Use e.g. the Finite Element
Method programme ANSYS. For the calculation of K ∗ , G∗T , G∗T,45 use Structural
problem, element type PLANE 82, element size 0.015, plane strain. For the
calculation of G∗L use thermal problem, element type PLANE 77, element size
0.015.

209
Part V
Fabrication processes
In this part we will study some of the fabrication processes concerning the mate-
rials and structures studied in Part I. In particular we consider the processes of
fabricating polymers, plastics and composites. Moreover, we study some of the
advanced manufacturing techniques of layer manufacturing technology (LMT).
From [4], we have the rough definition of what a material processing is: ”all
that is done to convert stuff into things”. Figure 21.1 gives an picture of the
different general materials processing types (independent of the material types).

Fabrication of an acceptable product involves the selection of both

i) an appropriate material and
ii) a companion method of processing,
such that the resulting combination can provide the desired shape, properties,
precision and finish.

22. Fabrication of plastics

The fabrication method depends on what raw material we have. Polymers are
formed by many low temperature processes. It is important to know for instance
whether the material is a thermoplastics or thermosetting. Thermoplastics can
be heated and then formed by either a formable solid or a liquid. Thermoset-
tings have far fewer options, because once the polymerization has occurred, the
framework structure is established and no further deformation can occur. Thus
the polymerization and the shape-forming are usually accomplished at the same
time. We will have a closer look at some of most used fabrication processes:

Casting
In casting, no fillers or no pressure is required. The liquid polymer is poured
into a container having the shape of the desired part. It is an inexpensive process,
dimensional precision is quite high, but quality problems can occur due to inade-
quate mixing, air entrapment, gas evolution, and shrinkage.
Typical products: sheets, plates, rods and tubes as well as small objects such
as jewelry.

210
Figure 21.1: Materials processes.

211
 Figure 22.1: Pressure die casting.

Pressure die casting

Pressure die casting (also called compression molding) is used for net-shape
forming (dimensional accuracy and surface details) of high-quality parts with com-
plex shapes. Thanks to the high pressures involved, very thin walled parts can be
obtained, see Figure 22.1. Preformed thermosettings (now also thermoplastics)
are introduced into an open, heated cavity. Pressure are applied. Products are:
gaskets, seals, exterior automobile panels, aircraft fairings , and a wide variety of
interior panels.

Blow molding
In blow molding, thermoplastics are converted into bottles or other hollow-
shape containers. The melted polymer is put into a mold, then compressed air
is used to spread the polymer into the mold, see Figure 22.2.It is used to make
many containers such as plastic soda containers and milk jugs.

Cold molding
In cold molding, raw thermosetting is pressed to shape while it is cold. Cured
in a separate oven. It is a low cost process, but neither surface finish nor dimension
precision is good.

Injection molding

212
 Figure 22.2: Blow molding.

Injection molding is the most widely used process for high-volume production
of thermoplastics parts. The resulting form is usually a finished product, no need
for other work before assembly or use. If the process is applied to thermosettings,
the process must be modified to provide the temperature and pressure required
for curing (which is additional time in a heated mold).
Similar to extrusion, the polymer is heated to the liquid state, but it is prepared
in metered amounts, and the melt is forced into a mold to create the part, see
Figure 22.3. This is not a continuos process. Many toys are made by injection
molding.

Transfer molding
Transfer molding combines elements from compression and injection molding.
Preformed raw thermosetting material is placed into a cavity, where it is heated
until molten which is forced into the channels in the die. Thin sections can
be made, with excellent detail, and good tolerance and finish. Inserts can be
incorporated into the product.
Products: switchgear and wiring devices, household appliances that require
heat resistance, structural parts that require hardness and rigidity under load.

213
 Figure 22.3: Injection molding.

Reaction injection molding

In reaction injection molding, two or more highly reactive liquid monomers are
mixed and immediately injected into the mold cavity where the chemical reactions
leading to solidification, occur, see Figure 22.4.
Heating is not required. Polyurethane (thermosettings urethane-groups), polyamides
(thermoplastics amids-group), composites with short fibers and flakes can be
molded. Different formulations result in elastomeric or flexible, structural foam
(foam core with a hard, solid outer skin), solid (no foam core), or composite prod-
ucts. The shapes can be quite complex (with variable wall thickness), and surface
finish is excellent.
Products: automobile products; steering wheels, instrument panels, door pan-
els, armrests, household; refrigerators, water-skis.

Insert molding
Insert molding is an injection molding process whereby plastic is injected into
a cavity and around an insert piece placed into the same cavity just prior to
molding. The result is a single piece with the insert encapsulated by the plastic.
The insert can be made of metal or another plastic.
Like injection molding in general, insert molding can be accomplished with
a wide variety of materials, including polyethylene, polystyrene, polypropylene,

214
 Figure 22.4: Reaction injection molding.

polyvinyl chloride, thermoplastic elastomers, and many engineering plastics. The

primary factors that restrict the use of insert molding are not process related,
but are determined by the strength and other properties required for the molded
product.
Processing Parameters: One of the chief causes of failure in an insert-
molded part is the cleanliness of the insert. It is absolutely imperative that the
insert be as clean as possible prior to molding. When molding with large metal
inserts, the inserts may need to be preheated to minimize the stresses caused by
differential thermal expansion and contraction. When inserts are manually loaded
it is important that the operator maintain a consistent cycle time.

Thermoforming
In thermoforming, thermoplastic sheet material is heated to a working tem-
perature and then formed into a finished shape by heat, pressure, or vacuum, see
Figure 22.5. There are two main steps in the process: heating and forming.
Products: range from panels for contoured skylights, large items such as bath-
tubs, to pages of text for the blind.
The size of Thermo Pressure Formed parts is limited by the size of the thermo-
forming machines now in use and the size of plastic sheet stock which is available.

215
 Figure 22.5: An example of thermoforming.

Commercially available sheet stock is attainable in sizes of 120 inches wide and
as long as required in thicknesses of up to .500 inches. Larger, thicker sheet stock
is available in special cases.
To date, the largest commercially available Thermo Pressure Formed parts are
in the size range of 48 inches long by 24 inches wide and 18 inches deep. However,
the size capabilities of the process increase almost daily.

Rotational molding
Rotational molding produce hollow, seamless products of a wide variety of
sizes and shapes, including storage tanks, bins, and refuse containers, doll parts,
footballs, helmets, and even boat hulls. Gravity is used inside a rotational mold
to achieve a hollow form, see Figure 22.6. The polymer is heated and due to the
rotation, distributed in the form of a uniform thickness all over the mold
The process is principally for thermoplastics (thermosets and elastomers are
more and more common).

Extrusion
By extrusion, long plastic products with uniform cross section are produced.
Thermoplastic pellets or powder are the raw material. The polymer is heated to
the liquid state and forced through a die under pressure resulting in an endless
product of constant cross section, see Figure 22.7. Extrusion is a cheap and rapid
method.
Products: tubes, pipes, window frames, sheet and even coated wires and cables.
60% of polymers are prepared in this way.

216
 Figure 22.6: Rotational molding.

Film Blowing
Film blowing uses the same method as extrusion: the material coming out of
the die is blown into a film. An example is plastic wrap.

Foam molding
In foam molding, plastic foam is fabricated. Foaming agent is mixed with
the plastic resin and releases gas or volatilizes when the combination is heated
during molding. The materials expand to 2-50 times their original size. The
process produce open-cell foams and closed-cell foams. Open cell-foams have
interconnected pores that permit the permeability of gas or liquid. Closed-cell
foams have the property of being gas- or liquid-tight. Both thermoplastics and
thermosettings can be foamed.
Polyurethane is one of the most versatile polymers in use today. It’s chemistry
is based on the reaction of an isocyanate with a polyhydroxy compound (polyol).
Therefore, a polyurethane foam-in-place system is a combination of the above
two components which have been preformulated so they need to only be mixed
and dispensed to make polyurethane foam, see Figure 22.8. North Carolina Foam
Industries (NCFI) was awarded a Mission Success Medal on July 29, 1998 by
Lockheed Martin Michoud Space Systems for production and delivery of a light-
weight spray-applied foam insulation system that Lockheed Martin applies on the
surface of the new super lightweight external fuel tank.

217
 Figure 22.7: Extrusion.

Figure 22.8: Foaming in-place. From: http://www.ncfi.com/foam-in-place.htm

Products: Rigid foams are useful for structural applications, packaging, ship-
ping containers, interiors of thin-skinned metal components, and flexible foams
are primarily used for cushioning.

22.1. Processing of Rubber and elastomers

The simplest of the fabrication processes is dipping, where a master form of metal
is first produced, which is dipped into the molten liquid of elastomers.
Dipping produce relatively thin parts with uniform thickness, such as boots,
gloves, balloons, swim caps, toys, and fairings.
Adaptions of the previously discussed processes for plastics are also used to
produce the shapes (injection, compression, and transfer molding, and extrusion).

218
Figure 22.9: Glovedipping, picture from: http://www.accautomation.com/prodsysc.htm.

References:
Fabrication Processes:
http://islnotes.cps.msu.edu/trp/toc.html
http://www/designinsite.dk
Fabrication of plastics:
http://www.mse.cornell.edu/courses/engri111/polymer4.htm
Foam molding:
http://www.ncfi.com/foam-in-place.htm
Videos:
http://www.gwplastics.com/capabilities/amto.html
3D Body scanner, see http://www.tc2.com/RD/RDBody.htm
Design
http://www.machinedesign.com/
Application of plastics:
http://www.stug.com.au/Applications/Common.html

Questions
• What kind of main groups, subgroups and processes of fabrication processes are
there?
• How does the fabrication of a thermoplastic polymer differ from the processing
of a thermosetting polymer?

219
• What are some of the ways that plastic sheets, plates and tubes can be fabri-
cated?
• What types of polymers are most commonly blow molded?
• What type of products are produced by rotational molding?
• How do we produce long plastic products with uniform cross-section?
• How are foamed plastics produced?
• What is the difference between open-cell foams and close-cell foams?

220
23. Fabrication of Fiber-Reinforced Composites (FRC)
Taking composite materials as a whole, there are many different material options
to choose from in the areas of resins, fibres and cores, all with their own unique
set of properties such as strength, stiffness, toughness, heat resistance, cost, pro-
duction rate etc. However, the end properties of a composite part produced from
these different materials is not only a function of the individual properties of the
resin matrix and fibre (and in sandwich structures, the core as well), but is also a
function of the way in which the materials themselves are designed into the part
and also the way in which they are processed. In this section we compare a few
of the commonly used composite production methods.

23.1. Manufacturing processes of Reinforcements

Individual fibres or fibre bundles can only be used on their own in a few processes
such as filament winding. For most other applications, the fibres need to be
arranged into some form of sheet, known as a fabric, to make handling possible.
Different ways for assembling fibres into sheets and the variety of fibre orientations
possible lead to there being many different types of fabrics, each of which has its
own characteristics. These different fabric types and constructions are explained
later.
Glass filaments are supplied in bundles either strands, rovings or yarns. A
strand is a collection of more than one continuous glass filament. Roving generally
refers to a bundle of untwisted strands wound in parallel to form a cylindrical,
flat-ended package, similar to thread on a spool. Single-end roving contains only
one continuous strand of multiple glass filaments. Multiple-end roving contains
numerous wound strands. Yarns are collections of filaments or strands (usually
smaller than rovings) that are twisted together.
Carbon Fiber: Fiber produced by carbonizing precursor fibers based on PAN
(polyacrylonitrile), rayon and pitch to eliminate non-carbon atoms. The term car-
bon fiber is often used interchangeably with graphite. Carbon fibres are produced
by the PAN or the pitch methods.
Pitch method pulls out graphite threads through a nozzle from hot fluid
pitch.
The PAN method separates a chain of carbon atoms from polyacrylnitrile
(PAN) through heating and oxidation. The polymer is stretched into alignment
parallel with what will eventually be the axis of the fiber. Then, an oxidation

221
 Figure 23.1: Woven mat.

treatment in air between 200 and 300◦ C transforms the polymer into a nonmeltable
precursor fiber. This precursor fiber is then heated in a nitrogen environment. As
the temperature is raised, volatile products are given off until the carbon fiber is
composed of at least 92% carbon. The temperature used to treat the fibers varies
between 1000◦ C and 2500◦ C depending on the desired properties of the carbon
fiber.
Aramid: A high strength, high stiffness fiber derived from polyamide. Kevlar ° R

and Nomex °
R
are examples of aramids.
Fiberglass: Filaments made by drawing molten glass, commonly used to reinforce
composite materials.
Woven mats: For applications where more than one fibre orientation is required,
a fabric combining 0◦ and 90◦ fibre orientations is useful.
Woven fabrics, see Figure 23.1, are produced by the interlacing of warp (0◦ ) fi-
bres and weft (90◦ ) fibres in a regular pattern or weave style. The fabric’s integrity
is maintained by the mechanical interlocking of the fibres. Drape (the ability of
a fabric to conform to a complex surface), surface smoothness and stability of a
fabric are controlled primarily by the weave style.
Filament winding: This is the automated process of wrapping resin impreg-
nated filaments, see Figure 23.2, (rovings or tows) in a geometric pattern over a
rotating male mandrel. The component is then cured under high pressure and
temperature.
Rovings: Rovings consist of one or more glass strands made up of varying num-
bers of continuous glass fibers of a specific filament diameter. These fibers are
coated with a sizing designed to protect the fiber during processing and to couple
with the customer’s resin to optimize laminate performance.

222
 Figure 23.2: Filament winding rovings.

Rovings, the most common form of glass, can be chopped, woven or processed
to create secondary fiber forms for composite manufacturing, such as mats, woven
fabrics, braids, knitted fabrics and hybrid fabrics. Rovings are supplied by weight,
with a specified filament diameter. The term yield is commonly used to indicate
the number of yards in each pound of glass fiber rovings.
Mats: Mats are nonwoven fabrics that provide isotropic or equal strength in
all directions. They come in two distinct forms: chopped and continuous strand.
Chopped mats contain randomly distributed fibers cut to lengths typically ranging
from 1.5 to 2.5 inches and held together with a chemical binder. Inherently weaker
than continuous-strand mats, chopped-strand mats provide low-cost polymer re-
inforcement primarily in hand layup, continuous laminating and some closed-
molding applications.

Chopped Strand Mat: The chopped strand mats, see Figure 23.3, are made
from cut fibers laid in a random pattern and bonded with a powdered, highly
soluble resin binder.
Continuous Filament Mat: In a different production step, strands formed
below the bushings are treated with a binder and formed into a swirl pattern to
make continuous filament mat.

23.2. Prepregs, Preforms and Compounds

A prepreg is an intermediate composite form made by preimpregnating the rein-
forcement with resin prior to molding. Molding compounds prepared in this way

223
 Figure 23.3: Chop strand mat.

are mostly glass fiber mats or glass filament cloths processed to form molded parts
or semi-finished products by hot-press molding
Preforming is an intermediate molding process whereby the reinforcement is
assembled in the shape of the part to be molded. This helps to ensure uniform
properties in a composite product and speed the molding cycle.
Glass fiber-reinforced thermoplastic compounds are typically in the form of
pellets consisting of resin with fibers in it. They can have several different forms,
i.e. sheets and billets/bulk form, also called bulk molding compounds (BMC) and
sheet molding compound (SMC).
The pre-mixed material has the consistency of molding clay and is injected
into a mold cavity or placed on the bottom of two mold halves for compression
molding.

23.3. Fabrication processes of FRC

There are several processes that manufacture fiber-reinforced composites. Some
of the processes are almost the same as we studied for manufacturing plastics, but
somewhat modified in order to incorporate the reinforcements. We will consider
some of the most widely used processes in manufacture FRCs.

Centrifugal Casting
This process makes cylindrical, hollow shapes such as tanks, pipes and poles.
Chopped strand mat is placed into a hollow, cylindrical mold, or continuous roving
is chopped and directed onto the inside walls of the mold. Resin is applied to the
inside of the rotating mold.

Cold press molding

224
 Cold press molding is a semi-open molding process. Reinforcements, usually
continuous filament mats (CFM), are placed into the tool and a highly filled
polyester resin is poured onto them. The press is closed and the part is cured.
The process requires lower pressures, 15 - 100 lb/in2, and temperatures averaging
55◦ C. Cycle times are in the range of 10 - 20 minutes. Other glass fiber fabrics,
mats, veils and preforms are also used. Parts usually have a fair-to-poor surface,
so uses are under-hood auto and truck components like fan shrouds, brackets and
battery supports.

Compression Molding (hot press molding)

The composite is formed under pressure between matching male and female
mold halves, at temperatures between 130 and 170◦ C. The material that is com-
pressed is typically one of several combinations of glass fibers and resin. These
combinations are: sheet molding compound, bulk molding compound, preform or
glass fiber-reinforced thermoplastic sheet.

Continuous Lamination
This is a process for making composite in sheet form such as composite glaz-
ing, corrugated or flat construction panels, and electrical insulating materials.
Reinforcement is combined with resin and sandwiched between two plastic carrier
films. The sheet takes shape under forming rollers, and the resin is cured to form
the composite.

Filament Winding
This process makes high strength, hollow and generally cylindrical products
such as pipe, storage tanks, pressure vessels and rocket motor cases. The fiber
is impregnated in a resin bath and pulled by the force of a rotating mandrel
which gives the part its shape, see Figure 23.4.Veils are used for inner and/or
outer surfaces to create a resin-rich surface for better corrosion resistance and
aesthetics.

Hand Lay-Up
This process is suited for making large, high strength parts at low to medium
volumes. A combination of reinforcements in roll form is laid into an open mold
and impregnated with resin. When the resin cures, the surface of the mold is
replicated on the side of the composite facing the mold.Resins are impregnated by
hand, see Figure 23.5, into fibres which are in the form of woven, knitted, stitched
or bonded fabrics. This is usually accomplished by rollers or brushes, with an

225
 Figure 23.4: Filament winding.

Figure 23.5: Hand lay-up.

increasing use of nip-roller type impregnators for forcing resin into the fabrics by
means of rotating rollers and a bath of resin. Laminates are left to cure under
standard atmospheric conditions.

Infusion molding
In infusion molding processes a single-sided mold which is covered with rein-
forcements and sealed with a flexible vacuum bag or film is used. A vacuum is
drawn on the space between the mold and the seal containing the reinforcements,
and a thermoset resin is allowed to infiltrate the reinforcements. The resin flows
through the reinforcements and cures to form the finished composite. Large parts
can be produced; such as boat hulls and windmill blades.

226
 Figure 23.6: Pultrusion.

Injection Molding
A thermoplastic or thermoset molding compound made of glass fibers and
resin is fed by a screw or plunger into a mold cavity. The mold halves are held
under pressure until the resin cures. Short glass fibers are commonly used as
reinforcements.

Pultrusion
This is a continuous process for making a lineal profile with a constant cross
section (such as rod stock, beams, channels and tubing) of fibre reinforced profiles.
After the reinforcement is impregnated with resin, the material is pulled through
a heated die that gives the part its cross sectional shape, see Figure 23.6.The resin
cures to create the composite profile.

Reaction Injection Molding (RIM)

This process uses a two-component resin system, see Figure 23.7. The two resin
components are combined and mixed together, then injected into a mold cavity
containing reinforcement. In the mold cavity, the resin rapidly reacts and cures to
form the composite part. The reinforcement is added to both resin components
before those components are reacted. The fiber-reinforced resin combination is
injected into the mold cavity where the resin rapidly reacts and cures to form the
composite part.

227
 Figure 23.7: Reaction injection molding.

RIM uses liquid raw materials unlike other methods (including injection and
compression) where plastic raw materials are in solid forms must first be melted,
then molded.

Resin Transfer Molding (RTM)

The reinforcement is placed in the bottom half of matching molds. After the
mold is closed and clamped, resin is pumped under pressure into the mold cavity.
The resin wets the reinforcement and cures to form the composite part.

Figure 23.8: Resin transfer molding.

Spray-Up
This is similar to and often combined with hand lay-up. With spray-up, glass
fiber roving is fed into a chopper ”gun” that chops the roving into fibers of pre-
determined length. The fibers are directed to a resin stream. The combination

228
 Figure 23.9: 3D woven sandwich fabric.

material is directed to the mold cavity when the composite part takes shapes.

23.4. Fabrication of Functionally Gradient Materials

Various techniques have been employed to the fabrication of FGMs, including
Chemical Vapor Deposition (CVD) / Physical Vapor Deposition (PVD), powder
metallurgy, plasma spraying, electro-plating and combustion synthesis.

23.5. Fabrication of sandwich constructions

3D woven sandwich
New methods makes it possible to braid, weave or knit sandwich structures,
as shown in Figure 23.9.The new research in this techniques and structures fo-
cuses on integrated core sandwich panels, based on 3D-Woven sandwich fabrics.
Intensive research on the 3D woven sandwich panels has been carried out, proving
the advantages of this material. The very high skin-core debonding resistance
of the panels is in theory very beneficial in increasing the lifetime and damage
tolerance of the structure. The application of 3D woven sandwich composites in
(semi-) structural composites depends on the long-term behavior and the damage
tolerance of the panels. These properties have not yet been investigated. Three
load cases are considered to be essential for the further applicability of the panels:
fatigue, impact and static concentrated loads.

229
1) 2)

glue
honeycomb material

3) 4)

Figure 23.10: Fabrication of aluminiun honeycombcore.

Manufacturing of honeycomb cores

There are two methods of constructing honeycombs; by Corrugation Process
or by Expansion Process.

• Corrugation Process

Sheets of the material are passed through corrugating rolls, to give them shape,
are then cut to length. The sheets are then bonded together to make a complete
honeycomb. This process only works for denser materials, which will retain their
shape after corrugation.

• Expansion Process

Sheets of the material are covered in thin strips of adhesive, and are stacked
to form a block, see Figure 23.10. The block is sliced to the right thickness, and
then pulled into an expanded form. This method is used for less dense materials.
Both of these methods produce a honeycomb in which the original material
runs in one direction. This is known as the web direction. The finished composite
material will be twice as stiff in the web direction as it is in the other.

Manufacturing of foams

230
 A new, inexpensive method for manufacturing cellular solids of highly repeat-
able and uniform open cell geometry has been developed. Pore sizes as small
as a few microns can be achieved. Materials choices include metals, ceramics,
glasses, polymers, semiconductors and composites. A significant advantage to
this method is that the distribution of material at the cell level is readily con-
trolled. This creates the possibility for optimally designed cellular structures with
exceptional thermophysical and mechanical properties. Multi-functional materi-
als are envisioned. These could include load bearing structures which also provide
for additional things such as impact/blast absorption, thermal management, con-
ductance of electricity and/or shielding of electromagnetic waves, electrical power
storage, filtering or impeding of fluid flow, retardation of chemical reactions and
fire or in-growth of biological tissue.
Thermoplastic foams are the result of deliberately adding at least one gas-
generating substance such as a chemical blowing agent, a soluble gas, or volatile
liquid under pressure to the polymer melt, then altering the environment to cause
the gas-generating substance to yield discrete bubbles. Foams offer unique process-
ing challenges, ranging from the technical aspects of controlling bubble nucleation
and growth to issues dealing with environmental concerns from diffusing blowing
agent gases.
Refractory foams can be fabricated from any material or materials combination
(either homogeneously combined or layered) which can be deposited by CVD
(chemical vapor deposition). Among the materials that can be deposited are the
refractory metals (e.g. niobium, tantalum, tungsten, rhenium) and their ceramic
compounds (e.g. the oxides, nitrides, carbides, borides, and silicides). Deposited
material densities of up to 50% of theoretical values can be readily achieved.
These materials can be furnished in various sizes and configurations, and are easy
to machine. Face sheets of either the same of different material can be applied to
enhance the flexural and tensile properties.
References:
Fabrication fiber-reinforced composites:
http://www.owenscorning.com/owens/composites/about/fabrication.asp
http://www.advancedcomposites.com/
http://www.giplastek.com/rim/
http://www.polymerprocessing.com/operations/rim/
http://www.raypubs.com/ctyp/ypoverview.html
Composite applications:

231
http://www.owenscorning.com/composites/applications/index.asp
About composites:
http://www.mdacomposites.org/materials.htm
http://www.netcomposites.com/
http://islnotes.cps.msu.edu/trp/toc.html
http://me.lsu.edu/~composit/
Manufacturing of foams:
http://www.uvapf.org/technology/viewInvention.cfm?inventionID=205
Fiber and Powder manufacturing
http://powdermetinc.com/
http://www.reade.com/home_index.html
http://www.wrigley-fibres.com/
http://www.claremontflock.com/
Design for Manufacturability
http://www.galorath.com/tools_manuf.shtm
http://www.npd-solutions.com/dfm.html
Design and analyse of composite materals:
Polymer Matrix Composites, http://mil-17.udel.edu/PMC/index.html
3D-woven sandwich:
http://www.mtm.kuleuven.ac.be/Research/C2/poly/PhD_hj.htm
http://www.muratec.net/braider/

Questions
• What is Prepreg?
• What is Preform?
• What is Compound?
• How is fiber manufactured?
• What are woven mats?
• How is carbon fiber manufactured?
• What different processes of making fiber-reinforced composites are there?
• How can honeycomb-cores be made?

232
 Figure 24.1: Layer Manufacturing Technology.

24. Layer Manufacturing Technology (LMT)

Layer Manufacturing Technology (LMT), is based on the principle of adding ma-
terial in 2D-layers to build a complete 3D-model, that is why it is called additive
technologies. Figure 24.1 shows how the parts made from LMT looks like in 2D
and 3D and close-up. There are several names of such processes: e.g. Rapid pro-
totyping (RP), 3D-printing, solid freeform fabrication (SFF), freeform fabrication
(FFF). The different technologies fabricate physical objects directly from CAD
data sources. These methods are unique in that they add and bond materials in
layers to form objects. Such systems offer advantages in many applications com-
pared to classical subtractive fabrication methods (for instance milling or turning),
such as

• Objects can be formed with any geometric complexity or intricacy without

the need for elaborate machine setup or final assembly.

• Objects can be made from multiple materials, or as composites, or materials

can even be varied in a controlled fashion at any location in an object.

• Solid freeform fabrication systems reduce the construction of complex ob-

jects to a manageable, straightforward, and relatively fast process.

• Small series of parts.

The latter has resulted in their wide use as a way to reduce time to market
in manufacturing. Today’s systems are heavily used by engineers to better un-
derstand and communicate their product designs as well as to make rapid tooling

233
to manufacture those products. Surgeons, architects, artists and individuals from
many other disciplines also routinely use the technology.
The names of specific processes themselves are also often used as synonyms
for the entire field of LMT. Each of these technologies has its singular strengths
and weaknesses.
LMT is an acronym for a group of processes capable of producing prototypes
of a complex geometry in various materials (wax, plastic, metal, etc.). The dom-
inating types of LMT processes, that will be studied in more detail, are:
• Ballistic particle manufacturing (inkjet)- BPM
• Fused deposition modelling - FDM
• Laminated object manufacturing - LOM
• Selective laser sintering - SLS
• Stereo lithography - SLA
• Solid ground curing - SGC (cubital)
• Three dimensional printing (3DP)
• Laser Engineered Net Shaping (LENS)
• Inkjets
• Powder metallurgy (PM)
Price is generally proportional to the time consumption on the machine. This
means that price increases with finer layer thickness, increased volume, as well as
increased part height.
The materials used in rapid prototyping are still pretty limited and dependent
on the method chosen. However, the range and properties available are grow-
ing quickly. Numerous plastics, ceramics, metals ranging from stainless steel to
titanium, and wood-like paper are available. At any rate, numerous secondary
processes are available to convert patterns made in a rapid prototyping process to
final materials or tools.
The ever increasing range of materials and processes in the group of additive
material processes make the processes possible to use in also fabrication of finished
products and not only prototypes. In for instance the SLS process, metallic powder
is the basis and the finished product has very good properties. Therefore, in
addition to prototypes, RP techniques can also be used to make tooling (referred
to as rapid tooling) and even production-quality parts (rapid manufacturing).
For small production runs and complicated objects, rapid prototyping is often the
best manufacturing process available. Of course, ”rapid” is a relative term. Most
prototypes require from three to seventy-two hours to build, depending on the

234
 Figure 24.2: Ballistic particle manufacturing.

size and complexity of the object. This may seem slow, but it is much faster than
the weeks or months required to make a prototype by traditional means such as
machining. These dramatic time savings allow manufacturers to bring products
to market faster and more cheaply.

24.1. Ballistic particle manufacturing (inkjet) BMP

BMP uses CAD-generated three-dimensional solid model data to direct streams
of material (waxes, plastics, photocurable polymers, ceramics, or metals) at a
target, building three-dimensional objects in much the same manner an ink jet
printer produces two-dimensional images. An object is built by a three-axis ro-
botic system controlling a piezoelectric ink-jet mechanism ”shooting” particles of
the material, producing multiple cross sections, onto a target, as shown in Figure
24.2.
There are different inkjet techniques (deposition systems), but all rely on
squirting a build material in a liquid or melted state which cools or otherwise
hardens to form a solid on impact.

24.2. Fused deposition modelling - FDM

FDM, see Figure 24.3 is another deposition method.The Titan machine from
Stratasys creates models using high performance engineering materials such as
polycarbonate, ABS and sulfones. Now with Titan, you can create prototypes
that have superior impact strength and also resist heat and corrosive agents such

235
 Figure 24.3: Fused deposition modelling.

as oil, gasoline and even acids. Multiple materials, coupled with one of the largest
build chambers in its class, make Titan a smart solution for building, large, strong
and durable parts. Parts up to 355 x 406 x 406 mm can be built.

24.3. LOM - Laminated object manufacturing

Laminated object manufacturing, see Figure 24.4 is a process suited for large and
heavy parts, e.g. models for metal castings. The process makes use of foils, where
the undersurface of this foil has a binder that when pressed and heated by the
roller causes it to glue to the previous foil. The foil is cut by a laser following the
contour of the slice. To help the removal of the excess material once the parts have
been built, the exterior of the slice is hatched, as opposed to fluid-based processes
(e.g. the SLA process), where the interior is hatched. Fine details cannot be
produced, and surplus material has to be removed manually from the part.
Helisys invented this process, but they ceased operation in 2000. However,
there are several other companies with either similar LOM technology, or in early
commercial stages. Maximum dimension was about 1600 * 1200 * 1200 mm.

24.4. Selective laser sintering - SLS

SLS, see Figure 24.5 produces part from plastic, wax, or metal powders. Advan-
tages include the use of industrial plastic types.

236
 Figure 24.4: Laminated object manufacturing.

Selective Laser Sintering is a thermal process and starts with a thin, evenly-
distributed layer of powder. A laser is then used to sinter (fuse) only the powder
that is inside a cross-section of the part. The energy added by the laser heats the
powder into a glass-like state and individual particles coalesce into a solid. Once
the laser has scanned the entire cross-section, another layer of powder is laid on
top and the whole process is repeated.
The hardened layer is lowered, a new layer of powder is spread, and the process
is repeated. Maximum dimensions are about 500 * 500 * 500 mm.
The SLS technology is known for its ability to process a variety of prototyping
materials including thermoplastics, investment casting wax, and a powdered metal
material for the production of prototype injection molds.

24.5. Laser Engineered Net Shaping (LENS)

Laser-engineered net shaping (LENSTM ) builds on the SLS process with a few
notable exceptions. Instead of bonding material in a bed of powder, the powder
is delivered in a gas jet through nozzles. Solid material is formed by solidification
of a molten pool, thus forming a fully dense, free-standing deposit.
A variety of materials can be used such as stainless steel, Inconel (a special
nickel material), copper, aluminum etc. Of particular interest are reactive mate-
rials such as titanium. Materials composition can be changed dynamically and
continuously, leading to objects with properties that might be mutually exclusive

237
 Figure 24.5: Selective laser sintering.

using classical fabrication methods.

The strength of the process lies in the ability to fabricate fully-dense metal
parts with good metallurgical properties at reasonable speeds.

24.6. Stereo lithography - SLA

Stereo lithography - SLA is the most common RPT technique, and produces
acrylic and epoxy parts from a liquid. Initially, an elevator is located at a distance
from the surface of the liquid equal to the thickness of the first, bottom-most layer,
see Figure 24.7. The laser beam will scan the surface following the contours of
the slice. The interior of the contour is then hatched using a hatch pattern. The
liquid is a photopolymer that when exposed to the ultra-violet (uv) laser beam
solidifies or is cured. The elevator is moved downwards, and the subsequent layers
are produced analogously. Fortunately, the layers bind to each other. Finally, the
part is removed from the vat, and the liquid that is still trapped in the interior is
usually cured in a special oven.
Almost unlimited geometry possibilities, fine geometric details and high accu-
racy.
The required geometry is produced in thin layers (0.0254 mm layer thickness)
that yields a smooth finish. A CNC-controlled laser beam cures a pattern in the
surface of a fluid photosensitive polymer. The hardened layer is then stepwise
lowered allowing the fluid to cover the part.

238
 Figure 24.6: Laser Engineered Net Shaping.

Typical applications are prototypes of injected moulded plastic parts and mod-
elling of human bones and skulls. Maximum dimensions are about 500 * 500 *
600 mm.

24.7. Solid ground curing - SGC

With the solid ground curing technique, see Figure 24.8, parts are produced from
photosensitive polymers by hardening them in layers just as the SLA process.
However, this is a significantly different process. The vat moves horizontally
as well as vertically. The horizontal movements take the workspace to different
stations in the machine.
The light source a uv-lamp (mercury) is used to flood the chamber and expose
and solidify the entire layer at once (instead of using a laser beam as in SLA).
This avoids the need for post-curing the parts. To select the areas that should
be cured, a mask is built on a glass plate, and subsequently, erased after begin
used. The mask is built using a process similar to the one used in laser printers.
The glass plate with the mask is placed between the lamp and the surface of the
workspace.
Parts are built surrounded by wax, eliminating the need for support structures
(which is needed in SLA). Once a layer has been exposed to the uv-lamp, the un-
cured areas-those areas filled with residual, liquid polymer-are replaced by wax.
This is done by wiping away the residual polymer and applying a layer of wax.

239
 Figure 24.7: Stereo lithography.

The wax is hardened by a cold metal plate, and subsequently, the layer is milled
to the correct height. The milling station also allows for layers to be removed,
i.e. an undo operation is possible. The new layer of polymer is applied when the
workspace moves from the milling station back to the exposure chamber.
The process is well suited for large and heavy parts. No separate part support
is required during processing. Compared with SLA, it is more expensive and not
as accurate. Machines from Cubital use SGC principle.

24.8. Three dimensional printing (3DP)

Three dimensional printing, see Figure 24.9 was developed at Massachusetts In-
stitute of Technology (MIT). It’s often used as a direct manufacturing process as
well as for rapid prototyping.
The process starts by depositing a layer of powder object material at the top
of a fabrication chamber. To accomplish this, a measured quantity of powder is
first dispensed from a similar supply chamber by moving a piston upward incre-
mentally. The roller then distributes and compresses the powder at the top of
the fabrication chamber. The multi-channel jetting head subsequently deposits a
liquid adhesive in a two dimensional pattern onto the layer of the powder which
becomes bonded in the areas where the adhesive is deposited, to form a layer of
the object.

240
 Figure 24.8: Solid ground curing.

Once a layer is completed, the fabrication piston moves down by the thickness
of a layer, and the process is repeated until the entire object is formed within
the powder bed. After completion, the object is elevated and the extra powder
brushed away leaving a ”green” object. No external supports are required during
fabrication since the powder bed supports overhangs.
Three dimensional printing offers the advantages of speedy fabrication and low
materials cost. In fact, it’s probably the fastest of all RP methods. Recently color
output has also become available. Parts can be made of any material (ceramic,
metals, polymers and composites).

24.9. PowderProcessing/PowderMetallurgy
Powder metallurgy is the process where both casting, forming and consolidation
are involved. Fine powdered materials are blended, pressed into a desired shape,
and then heated in a controlled atmosphere to bond the contacting surfaces of the
particles and establish the desired properties. Some of the advantages with the
process is that the material waste is very little, unusual materials or mixtures can
be used and the porosity or permeability can be tailored. There are many different
methods to manufacture products with powder metallurgy processes and in the
future even more methods will be introduced to the marked. Almost any metal,
metal alloy, or nonmetal such as ceramic, polymer or wax or graphite lubricant can

241
 Figure 24.9: Three dimensional printing.

be converted into powder and can thereby be produced by a powder metallurgy

method.
Powder metallurgy uses sintering process for making various parts out of metal
powder. The metal powder is compacted by placing in a closed metal cavity (the
die) under pressure. This compacted material is placed in an oven and sintered
in a controlled atmosphere at high temperatures and the metal powders coalesce
and form a solid. A second pressing operation, repressing, can be done prior to
sintering to improve the compaction and the material properties.
The properties of this solid are similar to cast or wrought materials of similar
composition. Porosity can be adjusted by the amount of compaction. Usually
single pressed products have high tensile strength but low elongation.
Powder metallurgy is useful in making parts that have irregular curves, or
recesses that are hard to machine. It is suitable for high volume production with
very little wastage of material. Secondary machining is virtually eliminated.
Typical parts that can be made with this process include cams, ratchets,
sprockets, pawls, sintered bronze and iron bearings (impregnated with oil) and
carbide tool tips.
FGMs can be fabricated by for instance Powder Metallurgy (PM).

242
24.10. Vapor Deposition CVD/PVD
Deposition is a process of depositing a thin layer of film onto the surface of the
wafer.
Vapor deposition processes usually take place within a vacuum chamber. There
are two categories of vapor deposition processes: physical vapor deposition (PVD)
and chemical vapor deposition (CVD). In PVD processes, the workpiece is sub-
jected to plasma bombardment. In CVD processes, thermal energy heats the gases
in the coating chamber and drives the deposition reaction.

Chemical Vapor Deposition is done by putting the chemicals into a heated

vacuum chamber containing a wafer. These chemical are fed into the chamber in
a gas form. There are a couple of different forms of chemical vapor deposition.
Each form has its own advantages and disadvantages. These forms are:
APCVD Atmospheric Pressure CVD
LPCVD Low Pressure CVD
PECVD Plasma Enhanced CVD

Physical vapor deposition does not use a chemical to deposit a film on the
wafer but uses physical means. To do this a gas inside a vacuum is excited with
a source , like RF (Radio frequency) energy. The molecules of the gas become so
energetic that if they were to collide with some materials they might cause atoms
of that material to break free. PVD uses this as a way to deposit layers of material
onto a wafer. A ”target” is placed above the wafer and when the molecules of gas
strike the target and cause the atoms to break free then a layer of atoms will fall
onto the wafer.

24.11. Selection of a layered manufacturing process

Selection of a layered manufacturing process depends on the basic application as
well as material and accuracy requirements. Selection rules of thumb have evolved:
Applications involving injection molded parts and/or good accuracy, often lead
to stereolithography (SLA) as the best choice.
If a prototype part needs to be in final materials and with mechanical proper-
ties that emulate injection molded parts, selective laser sintering (SLS) might be
chosen.
If the requirement is for large objects that will ultimately be sand cast, lami-
nated object manufacturing (LOM) is often the way to go.

243
 If the need is for a quick concept model and physical properties are of secondary
importance, three dimensional printing (3DP) may be most economical.
If the requirement is for extreme accuracy and very high resolution, such as
in making jewelry or small parts, a good choice is often an inkjet system such as
the ModelMaker Series from Solidscape, Inc. (formerly Sanders Prototypes).
Direct fabrication of metal objects as tools or prototypes will require selective
laser sintering (SLS), or one of the methods based on laser fusing such as laser
engineered net shaping (LENS).
In order to get the best finish of a product; rotate the geometry such that the
laser "draws" the 2D geometry that is most critical, and not let that surface be
step-wise.
Of course, such rules of thumb risk offending every system manufacturer be-
cause there is tremendous overlap in capabilities and specifications. The majority
of applications can probably be accomplished satisfactorily by several of the tech-
nologies.
References about additive technologies:
SLS- technology in action:
http://www.dtm-corp.com/
http://home.att.net/~castleisland/
http://lff.me.utexas.edu/sls.html
LMT Processes:
http://home.att.net/~castleisland/
http://designinsite.dk
http://www.efunda.com/processes/metal_processing/powder_metallurgy.cfm
http://www.efunda.com/processes/processes_home/process.cfm
http://www.efunda.com/home.cfm
http://www.precitech.ca/pmproc.htm
http://www.cs.hut.fi/~ado/rp/rp.html
http://home.att.net/~castleisland/rp_int1.htm
http://www-2.cs.cmu.edu/~rapidproto/manufacturing/mfgresources.html
http://home.att.net/~castleisland/rp_int1.htm
http://www-2.cs.cmu.edu/~rapidproto/manufacturing/mfgresources.html
http://www-2.cs.cmu.edu/~rapidproto/manufacturing/mfguniversity.html
http://emsh.calarts.edu/~mathart/R_Proto_ref.html
http://home.att.net/~castleisland/com_lks.htm
http://www.stratasys.com/ (also streaming video on FDM process)

244
Figure 24.10: Rapid Prototyping Technologies.

245
http://www.acceltechinc.com/nojava/sls.html
http://www.prototype3d.com/
http://www.3dsystems.com/
http://web.mit.edu/tdp/www/whatis3dp.html
http://www.myb2o.com/myb2ous/RapidPrototyping/Tools/Process/10265.htm
http://www.mit.edu/~tdp/ (pictures)= next link
http://web.mit.edu/afs/athena/org/t/tdp/www/
http://www.sandia.gov/media/lens.htm
http://www.tms.org/pubs/journals/JOM/9907/Hofmeister
/Hofmeister-9907.html (LENS)
CVD/PVD
http://www.corrosion-doctors.org/MetalCoatings/Physical.htm
http://entcweb.tamu.edu/zoghi/semiprog/deposit.htm
http://www.darpa.mil/dso/trans/fds_2.htm
Questions
• What are the different processes of LMT?
• What are the advantages and disadvantages of LMT-processes compared to
traditional processes?
• How does SLA work?
• Explain the powder metallurgy process

246
25. Design Considerations
The goals of reducing a large manufacturing cost and improving product quality
has lead to certain processes and procedures that have come to be known as design
for manufacturability or design for manufacture (DFM). The closely related is also
the area of design for assembly (DFA).

25.1. Designing for Manufacturability (DFM) )

Design for Manufacturability (DFM) or Design for Fabrication (DFF) is a method
and the theory for designing products that can be produced. In addition this
method saves time compared to previous methods (or non-methods).
In the past, products have been designed that could not be produced. Prod-
ucts have been released for production that could only be made to work in the
model shop when prototypes were built and adjusted by highly skilled technicians.
Clearly, there must be some other way to develop a product that is smarter than
this.
A product can be designed in many different ways. The designer’s objective
must be to optimize the product design with the production system. A company’s
production system includes its suppliers, material handling systems, manufactur-
ing processes, labor force capabilities and distribution systems.
Generally, the designer works within the context of an existing production
system that can only be minimally modified. However in some cases, the produc-
tion system will be designed or redesigned in conjunction with the design of the
product. When design engineers and manufacturing engineers work together to
design and rationalize both the product and production and support processes, it
is known as integrated product and process design. The designer’s consideration of
design for manufacturability, cost, reliability and maintainability is the starting
point for integrated product development.
A designer’s primary objective is to design a functioning product within given
economic and schedule constraints. However, research has shown that decisions
made during the design period determine 70% of the product’s costs while deci-
sions made during production only account for 20% of the product’s costs. Fur-
ther, decisions made in the first 5% of product design could determine the vast
majority of the product’s cost, quality and manufacturability characteristics. This
indicates the great leverage that DFM can have on a company’s success and prof-
itability.

247
 However, the application of DFM must consider the overall design economics.
It must balance the effort and cost associated with development and refinement of
the design to the cost and quality leverage that can be achieved. In other words,
greater effort to optimize a products design can be justified with higher value or
higher volume products.
Design effectiveness is improved and integration facilitated when:

• Fewer active parts are utilized through standardization, simplification and

group technology retrieval of information related to existing or preferred
products and processes.

• Producibility is improved through incorporation of DFM practices.

• Design alternatives are evaluated and design tools are used to develop a
more mature and producible design before release for production.

• Product and process design includes a framework to balance product quality

with design effort and product robustness.

25.2. Product design guidelines

Designers need to understand more about their own company’s production system,
i.e., its capabilities and limitations, in order to establish company-specific design
rules to further guide and optimize their product design to the company’s pro-
duction system. For example, they need to understand the tolerance limitations
of certain manufacturing processes.
DFM Guidelines:
• Minimize total number of parts
• Standardize components
• Use common parts across product lines
• Design parts to be multifunctional
• Design parts for ease of fabrication
• Avoid secondary operations
• Utilize the special characteristics of processes
In addition, colors, tolerances and materials also limits the process that should
be used.
Design guidelines

248
 There has been developed a number of rules for design. Briefly mentioned, the
list looks like this:
• Space holes in parts so they can be made in one operation without tooling
weakness. There is a limit in how close holes may be spaced due to strength in
the thin section between holes.
• Notes on engineering drawings must be specific and clear.
• Dimensions should be made from specific surfaces or points on the parts.
• Dimensions should all be from a single datum line.
• The design should aim for minimum weight consistent with strength and
stiffness requirements (this means minimized material cost and reduction in labor
and tooling costs).
• Adjust the design to the manufacture process chosen (wall thickness, fillets,
radii, dies etc.)
• Rotation of geometry

25.3. Evaluation of design alternatives

With the traditional approach, the designer would develop an initial concept and
translate that into a product design, making minor modifications as required
to meet the specification. DFM requires that the designer start the process by
considering various design concept alternatives early in the process. At this point,
little has been invested in a design alternative and much can be gained if a more
effective design approach can be developed. Only through consideration of more
than one alternative is there any assurance of moving toward an optimum design.
Using some of the previous design rules as a framework, the designer needs to
creatively develop design alternatives. Then alternatives are evaluated against
DFM objectives.
Design guide for composite materials, Part design concerns
The very nature of composites allows designers and manufacturers to tailor
fiber architecture to match performance requirements of a specific part. Fiber ori-
entation can be varied to allow wall-thickness variations, development of complex
shapes, and production of large parts with integral reinforcing members. Lam-
inates may be designed to be isotropic or anisotropic, balanced or unbalanced,
symmetric or asymmetric - depending on the in-use forces a component will en-
dure.
Understanding layered or laminate structural behavior is vital to designing
effective composite components. Adhesion of laminate layers - called plies - is

249
critical; poor adhesion can cause delamination under multiple stress, strain, im-
pact and load conditions. Ply layup designers must consider adhesion, strength,
weight, stiffness, operating temperature and toughness requirements, as well as
variables such as electromagnetic transparency and radiation resistance. Addi-
tionally, composite component design must encompass surface finish, fatigue life,
overall part configuration, and scrap or rework potential, to name just a few of
the many applicable factors.
The intended fabrication method will also affect design. For instance, manu-
facturers of filament-wound or tape-layed structures use different reinforcement
forms and build-up patterns than those warranted for hand laid up laminate
panels or vacuum-bag-cured prepreg parts. RTM accommodates 3-D preforms
more easily than do some other manufacturing techniques. The varying benefits
and limitations of these fabrication techniques provide considerable flexibility to
achieve part performance and economies.

25.4. The meaning of colors

Are Colors Significant? Only for the sighted or only if one isn’t color-blind. For
those of us blessed with sight, we’ve been taught that colors can make us feel
good, excite us, generate fear and joy, or literally make us nauseated. As long
as we attach a certain meaning to a particular color, the legends of colors will
continue to persist.
Color is a very important part to web design. As human beings we are very
sensitive to color. We rely heavily on certain colors to give meaning to things. For
example we associate the color red to love. Like hearts or roses. We also associate
red to an emergency like fire trucks or exit signs. As North Americans we associate
black to death however I bet you didn’t know that the Chinese culture associates
black with happy and joyful. We associate white to purity or wholesomeness like
a wedding dress, while in the Chinese culture a bride would never wear white
on her wedding day because it symbolizes bad luck or death. So you can see by
this simple example that color not only is symbolic for all cultures but NOT all
cultures associate color the same way.

How Are Colors Symbolic?

Time and tradition have created strong, symbolic color connections. For ex-
ample, we associate red with festivity, blue with distinction, purple with dignity,
green with nature, yellow with sunshine, pink with health, and white with purity.

250
Context also gives color specific leaning, we know that red means stop, and green
means go.
White: White is associated with peace and purity such as in the white clouds
and sparkling snowflakes. Two of the most popular peace symbols are the white
dove and the truce flag.
Red - Vitality, Courage, Self Confidence: When red is featured in a
dream the tint or shade carries different meanings. A light tint of red - suggests
the glory of success, hot reds - signify the stress of family quarrels, crimson hues
- foretell of happy news from a friend and red hair on a beautiful woman suggests
that you will receive unexpected good news.
Yellow - Wisdom, Clarity, Self-esteem: Studies show that yellow is most
often associated with words like cheerful, jovial and sunny, somewhat associated
with exciting and stimulating and almost never associated with words like respon-
dent, dejected, melancholy or unhappy. In short, a wonderful color for lifting the
spirits and letting the sun shine in.
Orange - Happiness, Confidence, Resourcefulness: The meaning of or-
ange is inexorably linked to the sensations of radiant energy, heat and the glowing
presence of the setting sun. The link between red and yellow, orange takes its
traits from both. It is less passionate and intense than red, incorporating the
sunny disposition of yellow. To the human eye, orange is seen as the hottest of
all colors, both in temperature and appearance.
Blue - Knowledge, Health, Decisiveness: The Russian artist Kadinsky
captured the expansive essence of blue when he wrote that blue is ”the infinite
penetration into the absolute essence - where there is, and can be, no end.” People
who are strongly attracted to blue tend to be devoted and deliberate in their
actions and since blue is everywhere, they feel a unity with the world.
Purple - Beauty, Creativity, Inspiration: The purple family is the most
enigmatic of all colors. Purple is a combination of the excitement of red and the
tranquillity of blue, the marriage of two diametrically opposed emotions.
Green - Balance, Love, Self control: Green is the most refreshing, restful
color to the eye. The word ”green” comes from the same root as ”grow,” so green
symbolizes that which grows: rebirth, regeneration, the renewal of life. Indian
mystics see green as the marriage of balance and harmony, the ray that bridges
cause and effect.

251
References:
Design guidelines:
http://www.devicelink.com/mddi/archive/96/04/008.html
http://www.arrem.com/designguide/dgdesvsmat.htm
http://www.tstar.com/pdf/quadrant-DF.pdf
http://www.npd-solutions.com/dfmguidelines.html
http://www.geplastics.com/webted/webted.html
Colors:
http://crumpledpapers.com/color.html
http://www.wevehadit.com/Publications/Color/page3.html
http://www.srsd.org/~arainone/color.html
Questions
• What is DFM?
• What is DFF
• What do we have to take into consideration when designing for fabrication?
• What kind of design-considerations would you take into consideration in general?
• How can colors affect our understanding of a product?
• If you would like a product to make you feel good, (happy), what color would
you choose on your product?

252
References
[1] Ashby M.F (1992), Materials Selection in Mechanical Design, Second edition,2000,
ISBN 0 7506 4357 9, Cambridge University Press, Cambridge.

[2] Alting Leo, (1994), Manufacturing Engineering Processes, ISBN-0-8247-9129-0, Se-

ries: Manufacturing engineering and materials processes, nr. 40, Marcel Dekker
Inc., New York.

[3] Cook R. D. , Malkus D. S. and Plesha M.E. (1989),Concepts and applications of

finite element analysis, J. Wiley and Sons Inc.

[4] DeGarmo Paul E. (1997), Materials and processes in manufacturing, New York,
Prentice Hall..

[5] Gibson L.J and Ashby M.F (1997), Cellular solids, Second edition, Cambridge
University Press, Cambridge.

[6] Gregg Bruce (1997), Modern materials and manufacturing processes, New Yok,
Prentice Hall.

[7] Greengard L. and Helsing J. (1998), On the numerical evaluation of elastostatic

fields in locally isotropic two-dimensional composites, J. Mech. Phys. Solids, 46,
1441-1462.

[8] Hashin, Z. (1983) Analysis of composite materials. A survey, J. Appl. Mech., 50,
483-505.

[9] Hashin, Z. and Shtrikman, S. (1962) A variational approach to the theory of effec-
tive magnetic permeability of multi phase materials, J. Appl. Phys 33, 3125-3131.

[10] Helsing J. (1995), An integral equation method for elastostatics of periodic compos-
ites J. Mech. Phys. Solids 43, 815-828.

[11] Hill R. (1964), Theory of mechanical properties of fibre-strengthened materials-I.

Elastic behaviour, Journal of the Mechanics and Physics of Solids, Vol. 12, 199-212.

[12] http://amsl.mit.edu/research/Devices/afc/AFCcompare.html

[13] http://www.chinanano.com/product.htm

[14] http://www.designinsite.dk/htmsider/inspmat.htm

253
[15] http://209.226.65.103/

[16] http://intellimat.com/

[17] http://www.zyvex.com/nano/

[18] http://www.cranfield.ac.uk/sims/materials/processing/fsm.htm

[19] http://www.ntu.edu.sg/home/mliuy/SMM.htm

[20] http://whirlwind.aem.umn.edu/people/faculty/shield/shp-mem.html

[21] http://www.plastics.com/definitions.php

[22] Lukkassen D. ,. Persson L.-E and Wall P. (1995), Some engineering and mathemat-
ical aspects on the homogenization method, Composites Engineering 5, 5, 519-531.

[23] Kelsey, S., Gellatly, R.A. and Clark, B.W. (1958) The sear modulus of foil honey-
comb cores, Aircraft Engineering, October, 294-302.

[24] Meidell, A. and Wall, P. (1997), Homogenization and design of structures with
optimal macroscopic behavior, in Hernández S. and Breddia C.A. (eds.), Com-
puter aided optimum design of structures V, Computational Mechanics Publica-
tions, Southampton, 393-402.

[25] Meidell, A. (1997), Layer manufacturing Technology, HiN-manuscript, Høgskolen

i Narvik, Narvik.

[26] Meidell, A. (1998), Anvendt Homogeniseringsteori, ISBN-82-7823-037-4, HiN-

report (51 p in Norwegian.), Høgskolen i Narvik, Narvik.

[27] Meidell, A. (1998), The out-of-plane shear modulus of two-component regular hon-
eycombs with arbitrary thickness, in Mechanics of Composite Materials and Struc-
tures (eds. C.A. Mota Soares, C.M. Mota Soares and M.J.M. Freitas), NATO ASI,
Troia, Portugal, Vol. III, 367-379.

[28] Meidell A. (1998), The out-of-plane shear modulus of two-component regular hon-
eycombs with arbitrary thickness, in Mechanics of Composite Materials and Struc-
tures (eds. C.A. Mota Soares, C.M. Mota Soares and M.J.M. Freitas), NATO ASI,
Troia, Portugal, Vol. III, 367-379.

254
[29] Meidell, A. (1999), Cellular Solids, HiN-report, Narvik Institute of Technology,
Narvik.

[30] Messler Robert W. Jr. (1993), Joining of Advanced Materials, ISBN-0-7506-9008-9,

Renesslaer Polytechnic Institute, Troy, New York.

[31] Persson L.E. ,Persson L. ,Svanstedt N. and Wyller J.(1993), The homogenization
method: An introduction, Studentlitteratur, Lund.

[32] Reddy J.N. (1984), Energy and variational methods in applied mechanics, John
Wiley and Sons, New York.

[33] http://en.wikipedia.org/wiki/Polymer#Morphological_Properties

[34] Wakil El and Sherif D. (1998), Processes and design for manufacturing, Boston,
PWS Publ. Co.

[35] Wright John R. (1996), Introduction to materials and processes, New or, Delmar.

[36] Zenkert Dan (1995), An introduction to Sandwich Constructions, Chameleon Press

Ltd., London.

255