4/1/15

AAE  590:  
Mechanical  Behavior  of  Materials  
Lecture  27:  Polymer  DeformaBon,  
Fracture,  and  Crazing  

DeformaBon  of  Polymers  

•  ElasBc  Behavior  
–  Higher  crystallinity  increases  elasBc  modulus  
–  ElasBc  deformaBon  via  stretching  and  bending  of  
primary  covalent  bonds  
•  PlasBc  Behavior  
–  Higher  crystallinity  greatly  increases  strength  
–  DeformaBon  via  complicated  processes  
–  Produces  highly  oriented  structure  

Figures  taken  from  Callister,  “Material  Science  and  

Engineering  An  IntroducBon”,  5th  Ed,  Wiley  (2000).  

1  
 4/1/15  

DeformaBon  of  Polymers  

•  Homogeneous  Shear  Flow  

•  Shear  Banding  
–  Localized  deformaBon  
–  Orient  along  direcBons  of  principal  shear  stress  
–  Necks  form  via  shear  bands  
–  Observed  in  tension  and  compression  
–  Primary  mechanism  in  amorphous  glassy  polymers  

 
Shear  Yielding  
Constant  Volume  DeformaBon  

Necking:  reducBon  in  sample  cross-­‐secBonal  area  

True  stress  ≥  Engineering  Stress,  necked  
region  must  be  stronger  (by  a  factor  of  λ)  
than  the  un-­‐necked  region  
Strain  Hardening  of  Polymers:  
 entropic  subchains,  crystal  orientaBon,  and  crystallizaBon  

2  
 4/1/15  

Polymeric  Behavior  by  Type  

Polymeric  Behavior  
•  Gels  
•  Elastomers  
•  Glasses  
•  Semicrystalline  
•  Large  anisotropy  of  bonding:  covalent  vs.  secondary   •  Fibers  
•  Homogeneous  vs.  heterogeneous  microstructures  
•  Microstructures:  amount,  size,  shape,  orientaBon,  and  connecBvity-­‐topology  
•  Large  (huge)  changes  in  microstructure  with  deformaBon  

 Strength/SBffening  Mechanisms  

S7ffening  Mechanisms   Strengthening  Mechanisms  

•  Increase  ElasBc  Modulus   •  Drawing  of  polymers  
–  Colder  temperatures   –  Analogous  to  cold  work  in  metals,  but  more  
dramaBc  
–  Higher  strain  rates   –  Strength  and  modulus  can  have  3x  increase  
–  More  crosslinking   in  drawing  direcBon  
–  More  secondary  bonding  
–  Higher  crystallinity  
–  Aligned  polymer  chains  
•  Decrease  DucBlity  
•  Heat  treatment  
–  Colder  temperatures   –  Prior  to  drawing,  may  lead  to:  
–  Higher  strain  rates   •  Higher  modulus  and  strength  
–  More  crosslinking   •  Reduced  ducBlity  
–  More  secondary  bonding   –  Post  drawing  
–  Higher  crystallinity   •  Reverse  strengthening  by  undoing  
–  Longer  chains  (higher  molecular   chain  orientaBon  and  strain  induced  
weight)   crystallinity  

3  
 4/1/15  

Glass  TransiBon  Temperatures  

 Temperature  at  which  polymers  transiBon  from  rubbery   Influence  of  
to  rigid  (glassy)  behavior   temp  on  σ-­‐ε  
–  Rubbery  above  this  temperature   behavior  of  
polymethyl  
–  Occurs  only  in  amorphous  regions  of  polymers   methacrylate    
–  Gibbs  free  energy  exceeds  the  acBvaBon  energy  for  
cooperaBve  movement  of  each  element  in  the  polymer  
–  Molecular  chains  slide  past  each  other  under  applied  force  

Figures  taken  from  Callister,  “Material  Science  and  

Engineering  An  IntroducBon”,  5th  Ed,  Wiley  (2000).  

Fracture  of  Polymers  

•  Fracture  strengths  of  polymers  is  low  relaBve  to  metals  and  ceramics  
•  Briele  Failure  
•  Apply  Linear  ElasBc  Fracture  Mechanics  (LEFM)  
–  Thermosefng  resins  
•  FormaBon  of  cracks  near  stress  concentrators  
•  Covalent  bonds  in  the  network  or  crosslinked  structures  are  severed  during  fracture   (*)  Briele  fracture  surface  of  
a  thermosefng  resin  
•  Crazing  
•  Proceeds  fracture  in  glassy  thermoplasBcs  (localized  yielding)  small  and  interconnected  microvoids    
•  Different  from  cracks  as  crazing  can  support  load  across  its  face  
•  Absorbs  fracture  energy  and  effecBvely  increases  the  fracture  toughness  
–  ThermoplasBcs  
•  Both  ducBle  and  briele  modes  are  possible  (dependent  on  glass  transiBon  temperature)  
•  Factors  that  favor  briele  failure:  lower  temp,  increase  strain  rate,  presence  of  sharp  notch,  increase  
specimen  thickness   (**)  
–  Amorphous,  glassy  polymers  
•  CavitaBons  
•  Light  refracBon  index  changes  (stress  whitening)  
–  Crosslinked,  glassy  polymers   (*)  DucBle  fracture  surface  of  a  thermoplasBc  
–  Crystalline  thermoplasBcs  
–  Rubbers  

Fibrillar   Microvoids   (***)  

Crack  
Bridges   (*)  Fracture  surface  of  a  polymer  fiber  
(***)  Callister,  Material  Science  and  Engineering  An   (*)  H.H.  Kausch,  Polymer  Fracture,  Springer-­‐Verlag,  1987.  
IntroducBon,  Wiley,  2000.  
(**)  J.G.  Williams,  Fracture  Mechanics  of  Polymers,  Wiley,  1984.  

4  
 4/1/15  

Craze  Yielding  
•  Crazing  in  tension  only,  not  in  compression  
•  ∆V>0  dilaBon/cavitaBons  –  void  formaBon  under  tension  
•  NucleaBon  at  flaws  –  usually  at  surface  
•  Crazing  only  occurs  for  (+)  hydrostaBc  tensile  stress  state  
•  Craze  plane  forms  perpendicular  to  max  principal  tensile  stress  
•  Voids  form  at  advancing  craze  Bp  due  to  dilataBonal  stresses  
•  Craze  Bp  growth  is  by  Taylor  meniscus  instability  of  fingering  and  fibril  pinch-­‐off  
•  As  bridges  elongate  and  break,  causing  microvoids  to  grow  and  coalesce,  which  leads  to  the  
formaBon  of  cracks  
•  Voids  grow  and  fibrils  elongate  as  polymer  chains  orient,  become  drawn  into  the  craze  and  work  
harden  
•  Craze  thickens  by  drawing  in  new  material  as  well  as  by  further  elongaBon  of  the  fibril  (decrease  in  
diameter)  

Craze  Microstructure  
•  Elongated  fibril/void  network  
•  Typical  void  content,  ~50%  
•  Fibril  diameter,  ~5nm    
•  Fibril  spacing,  ~20nm  
•  Draw  raBo  of  fibrils  is  1/(void  volume  fracBon),  ~2  
•  HeaBng  above  Tg  leads  to  healing  and  recovery  (entanglement  network  of  
subchains)  
•  Crazes  with  high  void  content  and  a  large  fibril  draw  raBo  are  prone  to  easy  
breakdown  (crack  propagaBon  and  briele  fracture)  
•  The  draw  raBo  (λ)  of  a  fibril  depends  on  the  contour  length  (l)  between  
entanglements  and  the  distance  between  entanglements  (d):  

l Mw
λ≈ ≈ ≈ Mw
d Mw
where  Mw  is  the  molecular  weight  of  the  polymer  

S.  Suresh,  FaBgue  of  Materials,  Cambridge,  2003.  

5  
 4/1/15  

Crazing  

Crazed  material  in  this  figure  has  a   Temperature  and  strain  rate  
residual  strain  of  ~60%  and  results  in   dependence  on  craze  iniBaBon  
permanent  deformaBon  
 
R.P.  Kambour,  A  Review  of  Crazing  and  Fracture  in  ThermoplasBcs,  Wiley,  1973    

Craze  Yielding  Criterion  

Material  Constants  (temperature  dependent)  

Ω1 (T )  Criterion  for  the  

σ max − σ min ≥ Λ(T ) + nucleaBon  of  a  craze  

Principal  Stresses  
σH in  a  3d  stress  state  
HydrostaBc  Stress  

Yield  Loci  for  Biaxial  Tension,  

Crazing  dominates  in  region  1  and   Yield  Loci  where  σ3<0  and  shear  
part  of  2  and  4   yielding  is  preferred  
Expanded  view  of  second  quadrant  where  σ1<0    and  σ2>0    

Pictures  from:  T.H.  Courtney,  Mechanical  Behavior  of  Materials,  McGraw-­‐Hill,  1990.  

6  
 4/1/15  

Crazing  vs.  Shear  Banding  

Both  result  in  localized  deformaBon  in  polymers  

1.  Shear  banding    

•  Constant  volume  deformaBon  
•  Oriented  along  direcBons  of  principal  shear  stress   Shear  
•  Occurs  in  compression  and  tension   Banding  
•  More  prevalent  in  semicrystalline  polymers  

2.  Crazing  
•  DilataBonal  deformaBon   Crazing  
•  Oriented  perpendicular  to  principal  tensile  stress  
•  Occurs  only  in  tension  
•  More  prevalent  in  amorphous  and  glassy  polymers  

 Crazing  vs.  Shear  Banding  

•  Shear  banding  occurs  if  the  number  of  entanglements  per  chain  is  large  so  
that  the  crazing  stress  is  higher  than  the  yield  stress.  
•  Crazing  occurs  if  the  yield  stress  is  lower,  for  example  if  the  molecular  
weight  is  low  (few  entanglements).    In  general,  lower  temperatures  favors  
crazing.  
•  In  many  cases,  heterogeneous  deformaBon  is  a  compeBBon  between  
crazing  and  shear  banding  in  polymers,  but  the  two  mechanisms  are  not  
mutually  exclusive.  

R.P.  Kambour,  A  Review  of  Crazing  and  Fracture  in  ThermoplasBcs,  Wiley,  1973     T.H.  Courtney,  Mechanical  Behavior  of  Materials,  McGraw-­‐Hill,  1990.  

7  
 4/1/15  

Crazing  vs.  Shear  Banding  

Role  of  drawing:  

–  Strengthening  Mechanism  
–  Orient  the  polymer  chains  
–  Increasing  strength  and  modulus  

R.W.  Hertzberg,  DeformaBon  and  Fracture  Mechanics  of  Engineering  Materials,  Wiley,  1989.  

Impact  Strength  
(*)  

•  Test  type:  Izod,  Charpy,  Plate  Impact,  etc.    

•  Some  polymers  have  briele  to  ducBle  transiBon  
•  Impact  strength  generally  increases  as  thermal  
souening  occurs  due  to  greater  ducBlity  

70
1045 Steel (**)  
Energy Absorbed during Impact (Joules)

60 7075 Aluminum
PMMA
50 ABS
(*)  
40

30

20

10

0
0 20 40 60 80 100 120
Temperature (deg C)

*  E.H.  Andrews,  Developments  in  Polymer  Fracture,  Applied  Science,  1979.  

**  R.P.  Kambour,  A  Review  of  Crazing  and  Fracture  in  ThermoplasBcs,  Wiley,  1973    

8  
 4/1/15  

Tear  Strength  
•  Ability to resist tearing
•  Energy required to tear apart a cut specimen that has a standard geometry
•  Related to tensile strength

•  Energy balance concepts of Griffith (brittle glasses)

•  Strain energy release rate, G:
1 d
G brittle = (Wext − U elastic )
t da
where a is the crack length, t is the thickness, W is the external work on the system, U is the elastic energy

•  Four energy sinks operative during crack propagation of crazed materials, in order of importance:
1.  Plastic work of craze formation
2.  Viscoelastic work of craze extension under stress
3.  Surface energy of the holes of the craze
4.  Breaking of chemical bonds

•  Griffith theory extended to tearing of rubbers by Rivlin & Thomas – 1952, at crack growth:

⎛ ∂Ω ⎞
G rubber = −2⎜ ⎟
⎝ ∂a ⎠ fracture

Grubber = strain energy release rate; tear strength related to energy required to produce unit area of crack, a
Ω = total energy of the system
•  Tearing of Rubbers ideally are ‘non-linear elastic’; little energy dissipation from the crack tip

E.H.  Andrews,  Developments  in  Polymer  Fracture,  Applied  Science,  1979.  

Environmental  DegradaBons  
•  Polymers  
–  Degrades  due  to  solubility  in  certain  liquids  or  absorbs  solvent  and  swells  
–  Crazing  leads  to  microvoids,  which  increases  suscepBbility  of  polymer  degradaBon  
from  liquids,  including  water!  
–  DegradaBon  due  to  alteraBons  in  molecular  structure  caused  by  radiaBon,  heat,  or  
light  source  
•  Polymeric  degradaBon  is  physiochemical  
•  Polymers  may  deteriorate  when  exposed  to  liquids  (water)  by  swelling  and  dissoluBon  
•  Swelling    
•  The  liquid  or  solute  diffuses  into  and  is  absorbed  within  the  polymer  forcing  the  
macromolecules  apart  and  causes  the  swelling  
•  Secondary  intermolecular  bonding  forces  are  reduced  making  the  material  
souer  and  more  ducBle  
•  DissoluBon    
•  Polymer  is  completely  soluble  in  the  liquid  
•  Both  are  affected  by  molecular  structure  and  temperature  
•  Mechanical  integrity  can  also  be  compromised  via  covalent  bond  scission  which  results  in  
a  reducBon  in  molecular  weight  
•  RadiaBon    
•  Chemical  reacBon  
•  Thermal  

9  
 4/1/15  

Case  Study:  Bone  Cement  

•  Also  known  as  poly(methyl  methacrylate)  

or    PMMA,  which  is  a  thermoplasBc  
•  Affixes  orthopedic  implants  and  remodels  
lost  bone  
•  Benefits:  
–  Biologically  compaBble  
–  Bone  cement  has  a  Young’s  modulus  
between  cancellous  and  corBcal  bone  
•  Promotes  load  sharing  
•  Avoids  stress  shielding  and  bone  
resorpBon  
–  Bone  cement  acts  like  a  grout  and  not  so  
much  like  a  glue  in  arthroplasty  

Mechanical  failure  of  bone  cement  is  linked  to  

failure  of  cemented  total  joint  prostheses!    
 

ApplicaBon  of  Bone  Cement  

•  Supplied  to  the  surgeon  as  a  powder  

and  liquid  (MMA)  and  mixed  
together  at  the  Bme  of  surgery  
•  Surgical  procedure:  
–  Remove  disease  Bssue  
–  Drill  out  marrow  cavity  and  fit  implant  
–  Mix  together  bone  cement  powder  and  
liquid  
–  Inject  cement  
–  Insert  implant  and  remove  excess  
cement  
–  Allow  cement  to  harden  or  cure  without  
moBon  

10  
 4/1/15  

Hardening  of  Bone  Cement  

 
•  Mixed  cement  is  liquid-­‐like  and  doughy  for  5-­‐10  
minutes  prior  to  hardening  
•  Hardening  reacBon  is  due  to  conversion  of  
monomer  to  polymer  (ie  in-­‐situ  free  radical  
polymerizaBon  –  iniBaBon  and  propagaBon  of  
polymer  chain)  
•  Bonding  is  mechanical,  not  chemical  (although  
sBcky,  it  does  not  bond  to  either  the  bone  or  
the  implant,  it  primarily  fills  the  spaces  
between  the  prosthesis  and  the  bone  
prevenBng  moBon)  

Porosity  in  Bone  Cement  

•  FormaBon  of  bubbles  
–  Bone  cement  heats  to  quite  a  high  temperature  while  sefng  and  
because  of  this  it  can  kill  bone  cells  in  the  surrounding  area  
–  Vaporized  monomer  may  end  up  in  blood  stream  or  cause  bubbles  
leading  to  voids  
•  PolymerizaBon  shrinkage  
–  Density  in  liquid  and  solid  state  are  1.18  g/cc  and  0.94  g/cc  
–  Full  conversion  of  monomer  to  polymer  results  in  21%  volume  
shrinkage  
–  Forms  pores  or  retracBon  from  cement/implant  or  cement/bone  
interface  
–  This  is  why  commercial  bone  cements  are  not  made  enBrely  of  
monomer  (other  reasons  include  temperature  rise  and  poor  handling  
properBes),  and  have  polymerizaBon  shrinkage  ~7%  
•  Effect  of  porosity  on  bone  cement  properBes  
–  Pores  significantly  weaken  bone  cement  
–  %  porosity  dependent  on  surgical  technique  
–  ConvenBonal  mixing  of  powder  and  liquid  results  in  many  pores  
–  CentrifugaBon  and  vacuum  mixing  result  in  lower  porosity  and  
significantly  improved  properBes  

11  
 4/1/15  

Microstructural  Pathway  of  Fracture  in  Bone  Cement  

Overview  (20x)  of  crack  and  pore  field  

Leading  crack  and  pore  field   Short  and  thin  microcracks   Leading  crack  and  microcrack  from   Open  leading  crack  face  
emanaBng  from  pores  (160x)   pore  (640x)   displaying  damage  zone  

Topoleski,  Ducheyne,  Cuckler,  Biomaterials  14  (1993)  1165.  

Fracture  ProperBes  of  Bone  Cement  

The  Effects  of  Mixing  Method  and  Molecular  Weight  
3-­‐Point  Bending  Fracture  
Toughness  Specimen  

Vacuum  Mixed  –  Less  Pores  

Hand  Mixed    

Fracture  Toughness  vs.   FaBgue  vs.  Mixing  Method   Fracture  Toughness  vs.  
Mixing  Method   Molecular  Weight  
Graham,  Pruie,  Ries,  Gundiah,  Journal  of  Arthroplasty  15  (2000)  1028.  

12  
 4/1/15  

Self  healing  polymers  

Figure  1  

•  SyntheBc  material  design  to  mimic  healing  response  of  nature  

•  Also  known  as  autonomous  or  smart  materials  
•  Invented  at  UIUC  by  co-­‐creates  Prof  Nancy  Soeos  (MatSE)  and  Scoe  
White  (AeroE)  
•  Tiny  capsules  containing  a  healing  agent  are  embedded  in  the  polymer.  
When  the  material  is  damaged,  the  capsules  rupture  and  release  the  
healing  agent,  which  repairs  the  cracks.  
•  A  microencapsulated  healing  agent  is  embedded  in  a  structural  
composite  matrix  containing  a  catalyst  capable  of  polymerizing  the  
healing  agent  (Figure  1).  
–  (i)  Cracks  form  in  the  matrix  wherever  damage  occurs.  
–  (ii)  The  crack  ruptures  the  microcapsules,  releasing  the  healing  agent  into  
the  crack  plane  through  capillary  acBon.  
–  (iii)  The  healing  agent  contacts  the  catalyst  triggering  polymerizaBon  that  
bonds  the  crack  faces  closed.  
•  SEM  image  of  the  fracture  plane  of  a  self-­‐healing  epoxy  with  a  
ruptured  microcapsule  (Figure  2).    
•  ApplicaBons  include:  Adhesives,  CoaBngs,  Biomaterials,  
Microelectronics,  AeronauBcal  Structural  Components,  Nanodevices,  
etc.  
Figure  2  

White,  Soeos,  Guebelle,  Moore,  Kessler,  Sriram,  Brown,  Viswanathan,  Nature  49  (2001)  794.  

 
Fracture  of  Self  Healing  Polymers  
SEM  of  crack  planes:  

Neat  epoxy  
(virgin  sample)  

crack  pinning  
toughening   Reference  
mechanism  
sample  with  
(10  wt%  
capsules)  

Crack  healing   Self  Healing  

efficiency,  η,  is  ability   sample    (10  wt%  
of  a  healed  sample  to   capsules  and  2.5  
Tapered  Double  
recover  fracture   CanBlever  Beam   wt%  catalyst)  
toughness   (TDCB)  Specimen  

Brown,  Soeos,  White,  Experimental  Mechanics  42  (2002)  372.  

13  
 4/1/15  

 
Self  Healing  Polymers  

Movie  from:  
hep://autonomic.beckman.illinois.edu/gallery.html  

14