BMFB 3263
Materials Characterization
Thermal Analysis
1
Thermal Analysis






A group of methods by which the physical & chemical
properties of a substance, mixture &/or reaction mixtures are
determined as a function of temperature and/or time, while
sample is subjected to a controlled temperature program.
Include heating or cooling (dynamic) or holding temperature
constant (isothermal), or combination.
Thermogravimetry Analysis (TGA) – mass of substance
against temperature or time.
Differential Scanning Calorimetry (DSC) – heat flow as a
function of temperature or time.
Thermal Mechanical Analysis (TMA) – deformation under
static load vs T or time.
Dynamic Mechanical Analysis (DMA) – substance under
oscillating load.
2
Thermogravimetric Analysis (TGA)


Used to measure changes in weight (mass),
m, of sample as a function of T and/or time.
Commonly used to




Determine polymer degradation temperature,
Residual solvent level,
Absorbed moisture content, and amount of
inorganic (noncombustible) filler in polymer or
composite material compositions.
Decomposition temperature of materials-impurities
in ceramic etc
3

Response ;


Weight gain – adsorption (physical), oxidation
(chemical).
Weight loss – vaporization (physical),
desorption (physical), oxidation (physical),
decomposition (chemical), dehydration &
desolvation (chemical).
4
Applications of TGA

Determines temperature and weight change of decomposition
reactions, which often allows quantitative composition analysis. May
be used to determine water content.

Allows analysis of reactions with air, oxygen, or other reactive gases
(see illustration below).

Can be used to measure evaporation rates, such as to measure the
volatile emissions of liquid mixtures.

Allows determination of Curie temperatures of magnetic transitions
by measuring the temperature at which the force exerted by a
nearby magnet disappears on heating or reappears on cooling.

Helps to identify plastics and organic materials by measuring the
temperature of bond scissions in inert atmospheres or of oxidation in
air or oxygen.

Used to measure the weight of fiberglass and inorganic fill materials
in plastics, laminates, paints, primers, and composite materials by
burning off the polymer resin. The fill material can then be identified
by XPS and/or microscopy. The fill material may be carbon black,
TiO2, CaCO3, MgCO3, Al2O3, Al(OH)3, Mg(OH)2, talc, Kaolin clay,
or silica, for instance.
5
Applications of TGA

Can measure the fill materials added to some foods, such as silica gels
and titanium dioxide.

Can determine the purity of a mineral, inorganic compound, or organic
material.

Distinguishes different mineral compositions from broad mineral types,
such as borax, boric acid, and silica gels.
Identify filler content, ash content by weight %.
Characterize decomposition patterns, study degradation mechanisms &
kinetics.
Prediction of lifetime (stability) at desirable time & temperature conditions
in desirable environment.
Screening of additives (stabilizers, flame retardant, plasticizers, etc).
Impurities analysis in materials.
Compositional analysis of polymeric formulations
Examine flame retardant & combustion properties.
Magnetic transitions in metallic materials.








6
Schematic thermobalance – sample heated at certain rate in a
controlled atm. Solid samples 1 mg to 100 mg but sometimes
up to 100 g.
7
8
TGA
9
Thermogravimetric Analysis




Sample is placed into a tared TGA sample pan which is
attached to a sensitive microbalance assembly.
Sample holder is then placed into high temperature furnace.
Balance assembly weigh the initial sample at room T & then
continuously monitors changes in sample weight (losses or
gains) as heat is applied to sample.
Heat applied at certain rate, in various environment. Typical
environment:
 ambient air, vacuum, inert gas, oxidizing/reducing gases,
corrosive gases, carburizing gases, vapors of liquids or
"self-generating atmosphere".
 The pressure can range from high vacuum or controlled
vacuum, through ambient, to elevated and high
pressure; the latter is hardly practical due to strong
disturbances.
10

Typical weight loss profiles are analyzed for
the amount or % of weight loss at any given
temperature, amount or % of noncombusted
residue at final temperature, & temperature
of various sample degradation processes.
11
(i) No decomposition with loss
of volatile products.
(ii) Rapid initial mass loss
characteristic of desorption or
drying.
(iii) decomposition in single
stage.
(iv) multi-stage decomposition.
(v) multi-stage decomposition
but no stable intermediates.
(vi) Gain in mass as a result of
sample reaction.
(vii) reaction product
decompose again.
12
Depending on polymer composition, reaction upon
heating will give their own characteristic TG curve.
Result can give thermal stability of material –
desorption, decomposition & oxidation information.
13
Calcium oxalate monohydrate – 3 distinct weight losses.
CaC2O4.H2O  CaC2O4  CaCO3  CaO
14
15
Molded underfill material (flip chip application) – 3 degradation
stages ; moisture & volatiles in resin, weakly bonded
monomers, then breakage of cross-linked monomers.
16
Multiple-stage reaction
: dehydration reaction
of hydroxide from
LiOH.H2O (exo).
4LiOH.H2O (solid) + O2
 2Li2O + 4H2O
Then formation
reaction of Li2SnO3
due to reaction
between Li2O with
SnO2 in mixture (exo).
17
• The sample was heated
from room temperature to
900°C at a rate of 5°C/min
in air.
•Polyester (71% of the
polymer),
•Polystyrene (29% of the
polymer),
• Thermogravimetric analysis was used as one of
several complementary techniques in the identification
of an unknown polymer composite.
• TGA was then performed on the material to find the
weight percent of each material.
• Fiberglass (22.9% of the
whole) and CaCO3 (49.3%
of the whole) were easily
identified by their different
temperatures of combustion
or evaporation.
•The combustion of the
styrene polymer component
produced enough energy
that the temperature
momentarily increased
more than the programmed
18
rate.
Thermogravimetric Analysis





Factors affecting TG curve
 heating rate
 sample size
 particle size of sample
 the way it is packed
 crucible shape
 gas flow rate
DTG – derivative of TG curve, often useful in revealing extra
detail.
TG also often used with DTA (differential thermal analysis).
DTA – record difference in T (∆T) between sample and
reference material. Each DTA curve should be marked with
either endothermic or exothermic direction.
Curve – peak represent exothermic or endothermic reaction.
19
What is endothermic and
exorthermic?

Endothermic- "within-heating" describes a process or
reaction that absorbs energy in the form of heat.



Comes from Greek prefix endo-, meaning “inside” and the Greek
suffix –thermic, meaning “to heat”.
E.g. melting of ice
Exothermic- one that releases energy in the form of
heat.
20
Differential Thermal Analysis (DTA)





Record temperature difference between sample & reference
material.
If endo event (e.g melting) temperature sample will lower than
reference material.
If exo event (e.g oxidation) response will be in opposite
direction.
Reference material:
 thermally stable at a certain temperature range
 Not react with sample holder or thermocouple
 both thermal conductivity
 heat capacity should be similar to those of sample
Both solid sample & reference material usually powdered form.
21
Butter & margarine – how different they are!
22
TG/DTA scan of
montmorillonite clay.
• Large endotherm at
114°C is assigned to loss
of interlayer absorbed
water.
• 2nd endotherm at 704°C
is dehydroxylation
reaction of the mineral.
• Last 2 peaks are
attributed to structural
changes, since no weight
loss are evident in TG.
23
Differential Scanning Calorimetry (DSC)





A thermal analysis technique in which the amount of energy
absorbed (endothermic) or released (exothermic) by a material
is measured.
Both events are the result of physical and/or chemical changes
in a material.
Normally the weight of sample is 5 – 10 mg,
Sample can be in solid or liquid form.
Many of the physical (e.g evaporation) or chemical (e.g
decomposition) transformation are associated with heat
absorption (endothermic) or heat liberation (exothermic).
24
DSC provides a direct calorimetric measurement of the
transition energy at T of transition. Often used to
characterize thermal transition in polymers – glass
transition T (Tg) and melting point (Tm). Organic liquids
or solids, and inorganic can be analysed.
25
Features of DSC curves
26
DSC – Applications :







Identify melting point, glass transition, Curie
temperature, energy required to melt material.
Evaluation of phase transformation.
Decomposition, polymerization, gelation, curing.
Evaluation of processing, thermal & mechanical
histories.
Process modeling, material’s min process temperature
(processing condition).
Determine crystallization temperature upon cooling.
Perform oxidative stability testing (OIT).
Compare additive effects on material.
27
Differential Scanning Calorimetry (DSC)





2 pans sit on a pair of identically positioned platforms
connected to a furnace by common controlled heat.
Record any energy difference – endothermic or exothermic
depending on whether more or less energy has to be supplied
to the sample relative to the reference material.
Endothermic response usually represented positive, opposite of
usual DTA convention (endothermic as negative side)
Correlate endothermic or exothermic peaks with thermal
events in sample.
One way – test if readily reversible on cooling & reheating.
Exothermic process usually not, unlike melting & many solidsolid transitions.
28
Computer makes sure that the 2 separate pans heat at
the same rate (usually 10°C/min or lower) as each
other. So if endothermic or exothermic events, results
in more or less energy has to be supplied to the
sample.
29
2 modes – depending
on method of
measurement used.
30
Heat absorbed by polymer, heat
flow given by q/t, q heat supplied
per unit time. Heat capacity, Cp
amount of heat it takes to
increase T. displacement, h =
BØCp. Ø is heating rate & B
calibration factor.
Heat is being absorbed by
sample (increase in its heat
capacity). Polymers gone thru
Tg, but transition occur over a
temperature range. So Tg is
taken as middle of the incline.
31
Crystallization point – Tc where at
this temperature polymer have
enough energy to arrange into
ordered arrangements, crystal.
Polymers give off heat at this point.
Area of peak = latent energy of
crystallization.
Heat absorbed in order to melt
– additional heat to increase
temperature. Area of dip = heat
of melting.
32
•Typical DSC curve for polymer (especially thermoplastic), for
polymers that don’t crystallize (amorphous), Tc & Tm will not
present.
•Comparing Tg with Tc & Tm, Tg only involve changes in heat
capacity.
33
DSC Responses


Physical changes :
 Exothermic – adsorption, crystallization.
 Endothermic – desorption, melting,
vaporization.
Chemical changes :
 Exothermic – oxidation, decomposition,
curing.
 Endothermic – reduction, decomposition,
dehydration.
34
DSC curve for typical organic
polymer.
Tg – change in heat capacity
but no change in enthalpy, ∆H =
0.
DSC directly
measures ∆H of
transitions. Also
degree of
crystallinity,
degree of curing.
35
During curing,
polymer chains
cross-link, this
process release heat,
which is reflected
DSC curve;
When polymer going
through curing
process, its DSC
curve will show a
broad peak at the
curing temperature
at first scan;
When cross-linking is complete, the curing peak
disappear and its replaced by a feature of glass
transition as shown in the curve of the second scan
36
Differential Scanning Calorimetry (DSC)





DSC have many applications in field of polymer
science & engineering.
Tg, Tc & Tm transitions are characteristic of each
polymer  identification.
Curing conditions for thermoset – heat for curing
which allows calculation of degree of curing.
But, DSC technology is not sensitive to detect Tg in
cross-linked or highly crystalline resins. Also for
polymer with high filler content.
Handling liquid also difficult. Interpretation of phase
transition requires further info – XRD, etc.
37
Polymer blend – immiscible blend. If fully soluble,
Tm peak will be in between Tm each elements.
38
% of crystallinity calculated relative to 100%
crystal material’s Tm peak.
39
Higher
crystallinity
Higher crystallinity gives larger & higher Tm peak.
40
Different grades gives different Tg, and thus,
different processing temperature.
41
42
THERMO-MECHANICAL ANALYSIS (TMA)





Dimensional properties of a sample are measured as sample is
heated, cooled or held under isothermal conditions. Loading or
force applied can be varied.
Change of dimensions as a function of temperature is recorded.
TMA measurements record changes caused by changes in free
volume of polymer.
Changes in free volume – by absorption or release of heat
associated with that change, loss of stiffness, increase flow,
change in relaxation time.
Free volume related to viscoelasticity, aging, penetration by
solvents, & impact properties.
43
As the space between
the chains increases,
the chain can move.
Physical aging
Tg
Increase in free volume caused by increased energy absorbed in
chains and this increased free volume permits various types of chain
movement to occur. Below Tg various paths with different free
volume exist depending on heat history & processing of polymer,
44
where the path with the least free volume is most relaxed.

Tg in polymer corresponds to the expansion of free
volume allowing greater chain mobility above this
transition.
45
Thermo-Mechanical Analysis (TMA)



Application :
 Determine coefficient of thermal expansion (CTE ) of
material.
 Identify Tg & Tm of material.
 Measure material’s heat deflection temperature (HDT).
 Composite delamination temperature.
All types of solid – powders, films, fibers, molded pieces, etc.
Use of probe resting on sample under a positive load. As
sample is heated, cooled or held isothermally, dimensional
changes in sample is translated into linear displacement of
probe.
46
Thermo-Mechanical Analysis (TMA)

Probe configuration – expansion, penetration,
compression, flexure, extension & dilatometry.
47
TMA : (a) penetration & (b) extension.
LVDT – linear variable differential transformer. Also
use other types of transducer – laser, optoelectronic.
48
TMA : (c) flexure & (d) torsional measurement.
Dimensional changes are monitored and transducer
transform responses into electrical signal (output).
49
Measurement of Tg of epoxy PCB – probe rest on
surface under low load. As sample expands during
heating, probe is pushed up & resulting expansion of
sample is measured. At Tg, epoxy matrix exhibits
significant change in slope due to an increase in its rate
of expansion. Onset T of this expansion = Tg.
50
CTE measurement – quantitative assessment of expansion
over a T interval. CTE before Tg = 50.5 µm/m°C, while above
Tg, CTE increases to 270.7 µm/m°C. Also shows residual
thermal stress – 1st heat result show undulation in region
near Tg  reflects release of stresses.
51
TMA penetration probe – during measurement, loading is
added so probe moves down thru sample as it softens.
Useful for measuring Tg of coatings on substrate. TMA
of wire sample with 2 coatings – inner coating prevents
electrical contact between adjacent wires and outer
coating used to bond the coil.
52
 TMA penetration results on crosslinked and noncrosslinked polyethylenes. Crosslinked sample exhibits
smaller degree of penetration due to higher viscosity in
liquid region above Tm.
 High sensitivity of TMA tech allows it to detect weak
transitions otherwise may not be observed by DSC.
53
Extensive crystalline transition, & softening point, Tg.
54
Another sample shows significantly smaller crystalline
to amorphous transition dimension increase compared
to 1st sample.
55
Thermo-Mechanical Analysis (TMA)




Since many materials are used in contact with
dissimilar material, rate & amount of expansion needs
to be known to help design around mismatches that
can cause failure of final product.
Limitation – only for solid samples. Also material
creep occurring concurrently with normal
dimensional changes.
Thermodilatometry – dimensional changes over wider
T range, up to 2000°C in variety of atm (inert,
vacuum, air, etc)
TD – sintering behavior of ceramic, clays.
56
Sintering behavior of kaolin & kaolinitic clays – on
heating loses water & form metakaolin structure.
Then converts to spinel structure (above 960°C) &
then above 1100°C to mullite.
57
DSC scan of amorphous metal alloy.
58
TMA of LDPE sample – compression mode
59
TMA of polyester partially oriented fibers –
extension mode
60
Dynamic Mechanical Analysis (DMA)


Characterize Visco-Elastic properties
Storage Modulus E’ (elastic response) and
Loss Modulus E’’ (viscous response) of
polymers are measured as a function of T or
time as the polymer is deformed under an
oscillatory load (stress) at a controlled
(programmed) T in specified atmosphere.
61
Dynamic Mechanical Analysis (DMA)





In viscous system, all work done by system dissipated
as heat.
Elastic system – work stored as potential energy.
Polymer  dual manner, i.e viscous-elastic.
DMTA – info on dynamic properties relating to these
2 behaviour. Gives 2 properties as a function of T :
 elastic modulus, E’ – energy stored (dynamic
storage modulus).
 viscous modulus, E’’ – ability to dissipate energy
as heat (dynamic loss modulus).
Thus, stiffness & its dampening capacity.
62
Dynamic Mechanical Analysis (DMA)





Measures dynamic modulus and/or damping of
material under oscillatory load as a function of T or
time at various frequencies.
Samples – fibers, films, molded sheets, powder.
DMA measures amplitude & phase of displacement
of sample in response to an applied oscillating force.
Data then used to calculate stiffness & convert to
modulus. Damping factor is also calculated.
A T scan at constant frequency can generate a
fingerprint of material’s relaxation processes & its Tg.
63
Dynamic Mechanical Analysis (DMA)




DMTA vs DSC – advantage of measuring side-chain
and main-chain motion in specific regions of polymer,
as well as relaxation.
Also, DTMA is more sensitive in detecting Tg
especially Tg of minor component.
The elastic modulus plotted as a function of T give
characteristic profile for polymer system.
Limitation – cannot measure mechanical properties
over full T range, because sample excessively
dampens the applied oscillation as it approaches its
softening point.
64
Dynamic Mechanical Analysis (DMA)




Sample is fixed between 2 parallel arms that are set
into oscillation by an electromagnetic driver at an
amplitude selected by operator.
DMA module measure changes in viscoelastic
properties of materials resulting from changes in T,
atm and time.
It then detects changes in the system’s resonant
frequency and supplies the electrical energy needed to
maintain the preset amplitude.
Frequency of oscillation is a measure of modulus of
the material. The amount of electrical energy needed
to maintain constant amp  damping properties.
65
An oscillatory strain is applied to sample in bending or
tensile deformation modes as a function of T or time.
Frequency & strain pre-selected & maintain constant.
66
DMA – sample subjected
to sinusoidally varying
stress of angular
frequency. For
viscoelastic material,
resulting strain will also
be sinusoidal, but will be
out of phase with the
applied stress owing to
energy dissipation as
heat, or damping.  is the
phase angle between
stress & strain. Damping
calculate from h = (v2 –
v1)/vr, or measure
driving force to maintain
constant amplitude. 67
Dynamic Mechanical Analysis (DMA)



Young’s modulus, E is related to square of resonance
frequency, r.
E = c L4  r2 / B2
c – constant, L – sample length between clamps, B –
sample thickness &  - sample density.
Damping usu expressed as logarithmic decrement per
cycle, i.e amplitude decay by half, log (A1/A2) = log 2
= 0.3. Damping is then 3.0 dB. Rate of decay is a
measure of how much damping is.
DMA output is plots of resonance frequency and of
damping as functions of T.
68
Dynamic Mechanical Analysis (DMA)


3 parameters are calculated : dynamic storage
modulus, E’, dynamic loss modulus, E’’, &
dissipation or damping factor, tan  = E’’/E’.
Application :
 examine viscoelastic behavior as a function of
stress, strain, frequency, time & T.
 measure modulus vs T.
 examine additive (filler, plasticizer) & cure time
effects on viscoelastic properties & Tg.
 examine mechanical behavior.
 quantify impact properties / toughness.
69
A – linear amorphous
polymers.
B – crosslinked
polymers.
C semicrystalline
polymers.
D & E – poly(ester
urethane)s.
# elastic modulus
plotted as a
function of T.
70
DMA result of FRP sample – resonant frequency vs T.
storage modulus is proportional to resonant frequency. As
T increased, resonant frequency decreased. On-set T
taken as Tg.
71
Transition phenomena / mechanical relaxation in HDPE. LowT relaxation, γ in polymers is normally associated with
improved toughness & impact behavior. α-relaxation is
assigned as Tg of polyethylene. α-relaxation requires
mobility within crystalline phase & coincides with
accelerated softening.
72
DMA of 2 diff samples of
PE. (a) 2 peaks : lower T
peak attributed to long
chain (-CH2-)n relaxation in
amorphous region, & higher
T peak to similar motion in
crystalline phase. T &
elative size can be related
to degree of crystallinity.
(b) Branched PE show 3
peaks, 1st and 3rd as above,
and 2nd peak is attributed
to –CH3 relaxation in
amorphous phase.
73
Behavior of styrene-butadienerubber (SBR). Various
formulations of SBR –
different styrene-butadiene
ratios, diff butadiene isomers,
diff additives, i.e carbon black
affected Tg, modulus of
elasticity.
# Changes in Young’s modulus
indicate changes in rigidity and
hence strength. Damping
measurements give practical
info on Tg, change in
crystallinity, occurrence of
cross-linking, & show up
features of polymer chains.
74
Study of thermoset
curing process (cure T &
time fixed) – amount of
cross-linker
(hexamethylene
tetramine) is shown to
effect not only the Tg,
but also modulus-T
behavior..
75