Sie sind auf Seite 1von 72

Special Report 88-9 US Army Corps

July 1988 of Engineers


Cold Regions Research &
Engineering Laboratory

Behavior of materials
at cold regions temperatures
Part 1: Program rationale and test plan

Piyush K. Dutta

DTIC
SEP6lg 8 -

Prepared tor88 0
OFFICE OF THE CHIEF OF ENGINEERS 8 10 6
Approved for public relcase; distribution is unlimited.
'II
UNCLASSIFIED
SECURITY CLASSIFICATION OF THIS PAGE
Form Approved
REPORT DOCUMENTATION "'AGE OMB No 0704 0788
_ Exp Date Jun30 1986
l1a REPORT SECURITY CLASSIFICATION lb RESTRICTIVE MARKINGS
Unclassified
2a SECURITY CLASSIFICATION AUTHORITY 3 DISTRIBUTION/AVAILABILITY OF REPORT
Approved for public release;
2b DECLASSIFICATION/DOWNGRAL)ING SCHEDULE distribution is unlimited.

4 PERFORMING ORGANIZATION REPORT NUMBER(S) S MONITORING ORGANIZATION REPORT NUMBER(S)

Special Report 88-9


6a NAME OF PERFORMING ORGANIZATION 6b OFFICE SYMBOL 7a NAME OF MONITORING ORGANIZATION
U.S. Army Cold Regions Research (If applicable)
and Engineering Laboratory CECRL Office of the Chief of Engineers
6c. ADDRESS (City, State, and ZIP Code) 7b ADDRESS (City, State, and ZIP Code)
Hanover, New Hampshire 03755-1290 Washington, D.C. 20314-1000

Ba. NAME OF FUNDING/SPONSORING 8b OFFICE SYMBOL 9 PROCUREMENT INSTRUMENT IDENTIFICATION NUMBER


ORGANIZATION (If applicable)

%. 8c. ADDRESS (City, State, and ZIP Code) 10 SOURCE OF FUNDING NUMBERS
PROGRAM PROJET TASK WORK UNIT
ELEMENT NO. NO 4A7627 NO ACCESSION NO
6.27.30A 30AT42 SS 019
11 TITLE (Include Security Classification)
'-.. Behavior of Materials at Cold Regions Temperatures
Part 1: Program Rationale and Test Plan

E 12 PERSONAL AUTHOR(S)
Piyush K. Dutta
13a TYPE OF REPORT 113b TIME COVERED 14 DATE OF REPORT (Year, Month, Day) 15 PAGE COUNT
FROM TO July 1988 72
16 SUPPLEMENTARY NOTATION

17 COSATI CODES 18. 4ACT TERMS (Continue on reverse if necessary and identify by block number)
F;ELD GROUP SUB-GROUP Cold regions Low temperatures
Compression~tension testing Materials testing
1e
19 ABSTRACT (Continue on reverse if necessary and identify by block number)
Newer materials and products are being constantly added to the Army's inventory. Cold regions climatic
• conditions should not impair the reliability and durability of these new systems. This report discusses
the rationale of the test program being undertaken at CRREL to evaluate material behavior at low
temperatures.

20 DISTRIBUTION /AVAILABILITY OF ABSTRACT 21 ABSTRACT SECURITY CLASSIFICATION

" UNCLASSIFIED/UNLIMITED El SAME AS RPT El DTIC USERS Unclassified


22a NAME OF RESPONSBLE IND:VIDUAL 22b TELEPHONE (Include Area Code) 22c OFFICE SYMBOL

Pivush K. Dutta 603-646-4100 CECRL-EA


DD FORM 1473, 84 MAR 83 APR edition may be used until exhausted SECURITY CLASSIFICATION OF THIS PAGE
All other erjiton ArP o UNCLASSIFIED

N..........................................................."- .",",..-.... ",. ,...


PREFACE

This report was prepared by Dr. Piyush K Dutta, Materials Research Engineer,
of the Applied Research Branch, Experimental Engineering Division,U.S. Army
Cold Regions Research and Engineering Laboratory. Funding was provided by
DA Project 4A762730AT42, Design, Construction and Operations Technology for
Cold Regions, Task SS, Service Support, Work Unit 019, Behavior of Materials at
-' "Low Temperatures.
The author thanks Dr. Ronald Liston of CRREL and Dr. Harold Lord of Michi-
gan Technological University for technically reviewing this report.
The contents of this report are not to be used for advertising or promotional pur-
poses. Citation of brand names does not constitute an official endorsement or ap-
proval of the use of such commercial products.

., .- *,.

- ..

S"
%-

0?r

"" ii

0".

.. . . .- . F. p . ".-.-
. - ~ .
CONTENTS
Page

Abstract ................................................................................. i
Preface ................................................................................. ii
Introduction ............................................................................ 1
*'. Low temperature properties of manufactured materials-background ........ 2
Metals .............. ................... ...................2
Plastics/polym ers ................................................................. 5
C omposite ........................................................................... 6
Fracture property tests for low temperature application of materials ........... 8
- Early fracture test approaches .................................... 8
Fracture m echanics test ........................................................... 11
Fracture toughness tests for polymers and composites ................ 16
Influence of cyclic load on fracture toughness ................................. 16
Strain rate effect on fracture ...................................................... 17
T est plan ............................................................................... 19
Rationale ........................................................................... 19
Experim ental program ............................................................ 19
Expected information ......................................... 25
* Selected bibliography ................................................................. 26
Appendix A: Low temperature behavior of some structural metals
and alloys ........................................................................ 31
Appendix B: Low temperature behavior of some selected polymers ............. 45

ILLUSTRATIONS

Figure
1. Photograph of T-2 tanker that failed at pier ........................ 2
2. Impact test results for a steel pipe material .................................. 3
3. Typical stress-strain curve of a body centered cubic class metal at
decreasing temperatures ............................................... 3
4. Simplified deformation behavior maps of FCC and BCC material ....... 4
5. Energy-temperature curves obtained by Charpy V-notch tests of
twelve steels .......................................... 4
6. Therinomechanical curve of a simple polymer ............................ 5
7. Schematic presentation of viscoelastic properties of polymers.......... 5
8. Stress-strain behavior of epoxy resins at different strain rates, i and
* tem peratures, T .......................................................... 6
9. Simple classification of composite system .................................. 7
10. Orientation dependence of composite property .............................. 8
11. Temperature dependence of composite property ............................ 8
12. Temperature effect on the fracture stress .................................... 9
13. Temperature effect on ductility ............................................... 9
14. Simple beam Charpy impact machine ....................................... 10
15. Centrally cracked panel under uniform axial stress ...................... 11
16. Plain strain and plane-stress fractures ..................................... 13
17. Typical load-crack displacement curve ......................... 15
18. Transition of fracture toughness ............................................. 15
19. Test program schematic ............................. ....... 20
* 20. Device for quasi-static slow and fast loading ............................... 20
21. Split Hopkinson pressure bar applied to fracture dynamics .............. 22
.. iii
', TABLES
Page

Table
1. Mechanical equipment and instrumentation requirements for low
temperature test of materials ........................................... 24

".J.
40 %

7iv

.. iv
4 Behavior of Materials at Cold Regions Temperatures
Part 1: Program Rationale and Test Plan

FIYUSH K DUTTA

INTRODUCTION

Cold regions engineering has a general need for the evaluation of materials at
low temperatures. Performance of many structures and components is seriously
affected when the weather becomes very cold. At low temperatures materials tend to
become hard and brittle; as a result, legitimate concerns are raised about their
reliability and safety in such cold weather. The disastrous accident of space shuttle
Challenger because of low-temperature-induced failure of O-ring seals is a grim
reminder of such problems. Moreover, newer and lighter materials and products
"-" .for engineering structures and systems are rapidly replacing the older and heav-
ier materials and components. There is a growing need to evaluate these materials
at low temperatures. In particular, the U.S. Army needs to evaluate new metal-
substitute polymers and polymeric composites in cold climates for its new High-
Technology Light Division.
Because of these needs, a low temperature (down to -60°C) material testing pro-
gram is being esttblished at CRREL. The testing program includes a variety of
tests ranging from ordinary compression-tension testing of specimens to loading
at very high 4train rates in a Hopkinson split-bar system. Destructive, non-
destructive and fracture mechanics tests are also planned to be performed at low
temperatures. This report discusses the rationale for this program and outlines the
test plan.
Cold regions material problems are many. Brittle-fractures initiated by low
temperature or high strain rate (or both) have at times caused large-scale damage.
In cold weather (00 to 50C) merchant vessels have broken in two while in harbor
(Fig. 1), bridges have collapsed, and pipelines and gas storage tanks have ripped
open (Parker 1957). Exploration and production of oil and gas have begun to take
place in colder climates (down to -60C), and in deeper and rougher water; these
rigorous conditions aggravate the brittle fractu-e hazard. At present, the knowl-
edge necessary to prevent brittle fracture is far from complete.
% Polymeric composites also have the potential for catastrophic failure at cold re-
gions temperatures. These materials tend to develop brittle failure characteristics,
depending on their composition and manufacturing processes. The development of
brittle failure in these materials accelerates with lower temperature, material de-
fects (notches or cracks), fatigue, and high-strain-rate loading.
In steel, a small reduction within a narrow temperature range called the transi-
tion temperature can cause a sharp decrease in fracture toughness. A preexisting
* crack that has become arrested at normal temperature and is thus tolerated can be-
come self-propagating at low temperature and may cause catastrophic structural
failure.
V, CRREL's low temperature material testing program will perform both quasi-
static and high-strain-rate loading fracture toughness tests on selected materials.
Tests will be done on a variety of materials, including plastics, composites, and
metals. Some materials will be subjected to cyclic loading at lower temperatures

.
;r,;

0Z
0 .. On

Figure 1. Photographof a T-2 tanker that failed at pier.

before the fracture mechanics tests are performed in order to evaluate property
degradation (fatigue) in service.
Quasi-static tests will be performed using the MTS and INSTRON material
testing machines now available at CRREL. A high-strain-rate loading facility us-
,W ing the Hopkinson split-bar system has been installed and suitably instrumented
to record and analyze the high-speed stress waveform data.
In the following sections the general background of low temperature material
behavior and critical physical and engineering property data are reviewed and
discussed. An outline of the proposed program for investigating the critical phys-
ical properties of a few selected materials at low temperature is also presented.

LOW TEMPERATURE PROPERTIES OF


MANUFACTURED MATERLIS-BACKGROUND

Most manufactured materials exposed to low temperatures show a substantial


loss of useful structural properties. Generally speaking, as the temperature is low-
ered, the hardness, yield strength and modulus of elasticity increase, but fracture
toughness/impact strength, fatigue strength, Poisson's ratio, thermal expansion
coefficient and specific heat decrease.

Metals
* At low temperature the most significant property change in metals is the in-
crease in brittleness. A metallic object or structure will shatter or fracture when
* subjected to stresses (especially from impact) that are allowable at normal temper-
atures (Fig. 2). The metals that have face-centered-cubic lattice structure, e.g.
nickel, copper, aluminum, lead and silver, show some ductility at low temperature.
But a larger group of body-centered-cubic class metals, e.g. iron, chromium, mo-
0.: lybdenum and tungsten, show a marked decrease in ductility.

2
.,*
0'
4r1

" 30 - % I

-BO -60 -40 -20 0 20 40 60 80


Rooi temperature Extreme cold Temperalure *F)

Figure 2. Impact test results for a steel pipe material.

200'F

0 -150

V)

Failure
-100
Figure 3. Typical stress-strain
70 curve of a body centered cubic
class metal at decreasing tem-
Strain - peratures.

Loss of ductility in a metal can be observed by examining its low temperature


stress-strain relationship. As the temperature is lowered, both the yield point
(where ductility begins) and the ultimate strength point (where failure occurs) may
shift to a higher stress value, but fracture may begin at a much lower strain value
(Fig. 3).
* Figure 4 shows the general aspects of temperature-dependent mechanical behav-
ior of both face-centered-cubic (FCC) and body-centered-cubic (BCC)metals. Note
that at the lowest stresses a specimen deforms elastically; increasing stress even-
V" tually brings the specimen to the limit of the elastic region where plastic strain
(permanent deformation) or brittle fracture occurs. Figure 4a shows the behavior of
FCC materials. Note that the yield strength increases with decreasing temperature,
• but the increase in ultimate strength is larger at lower temperatures. Figure 4b il-
lustrates the BCC behavior. This figure shows a brittle region where the specimen
Z' fails before any significant plastic deformation occurs. This brittle region is ex-
tended to higher temperatures by several factors, namely (1) sharp cracks or other
stress concentrators, (2) higher strain rate, and (3) impurities, e.g. higher carbon
0 contents in steel.

3
- 1 4
1040 1

20-
,,
3 Ductile
' 10- I CAC

80-
.,": 6 WC D

(1) Elastic Strain 40-

I __ ___ _ 20--
o Br~ftle [
a'lure- -120 -80 -40 0 40 80 120 160 200
,1" 1 Ductile -- Testing Temperoture (F)

' Composition
10 Steel C Mn Si P S Ni Cr

0 (2) A 0.17 0.53 0.25 0.011 0.02 3.39 0.06


(I B 0.18 0.82 0.15 0.012 0.03 0.04 0.03
Q2 C 0.18 0.73 0.07 0.008 0.03 0.05 0.03
TIT m , Normolized
D 0.20 0.33 0.01 0.013 0.02 0.15 0.09
Temperature

Figure 4. Simplified deformation be- Figure 5. Energy-temperaturecurves ob-


havior maps of (1) FCC materialand tained by Charpy V-notch tests of steels.
(2) BCC-material.

The fracture mode transition from relatively ductile to relatively brittle is be-
lieved to occur because of the very rapid rise in viscosity with decreasing tempera-
ture. In most materials the flow stress is low at high temperatures. The stress con-
centrations around a dislocation can always be relieved by local plastic deforma-
tion and there is no brittle fracture. At low temperatures, the yield stress for slip is
high so that deformation will start by microcrack production and subsequent brittle
fracture (McChntock and Ali 1966).
Because of the onset of brittle fracture, most standard constructional carbon
* steels cannot be depended upon at the temperatures encountered in cold regions.
Figure 5 shows energy-temperature curves obtained by Charpy V-notch tests of a
few steels having different chemical compositions. The curves clearly show the
temperatures where transition in the fracture mode occurs (McClintock and Ali
1966). However, continued research in steel development and processing has re-
sulted in a new class of high strength, low alloy steels that can be used in low tem-
perature environments (Baldy 1976).
The low temperature properties of a number of metals have been summarized in
% Appendix A. These properties indicate that stainless steel is probably the best-suited
ferrous metal for cold use. Unlike other steels, stainless steel has no transition
from tough to brittle in a cold environment. The properties of aluminum do not de-
0 grade at low temperature, and it is a preferred material for many cold weather ap-

.0,
plications. The ductility of cold-worked copper actually increases at low tempera-
ture, which makes it a very good metal for use in the cold regions.

Plastics/polymers
Plastic materials generally become more brittle at low temperatures, but they are
not as consistent as metals. The tensile strength of plastics increases very mark-
edly at lower temperatures. For example, a nylon material increases in tensile
strength from 7400 psi at 21'C to 13,000 psi at -57 0C, but its impact strength decreases
from 16 if-lb at 21 0C to 0.9 ft-lb at -40'C (Hansen 1960). Some plastics, for example
polyethylene, a thermoplastic polymer, remain tough at temperatures as low as
-73°C. However, serviceability of rubber components, e.g. tires, inner tubes, cable,
hose, bushings and seals, is seriously affected by low temperature. Rubber develops
brittleness and loses flexibility; its loss of resilience is associated with changes in
hardness, volume and coefficient of thermal expansion.
Both time and temperature influence the behavior of polymers. Figure 6 shows
the typical thermomechanical behavior of a simple polymer. At low temperature it
behaves as a glass. As the temperature is raised, it becomes less brittle. The upper
limit of the glassy region is called the glass transition temperature, Tg. Com-
monly observed values of Tg are in the range of-53C to -97 0 C (Read 1983). In the
glassy region, polymers typically have a modulus of elasticity of 105 to 106 psi if or-
ganic and up to 10 7 psi if inorganic. Poisson's ratios are 0.25 to 0.4. In the rubbery
region the modulus drops to 102 to 103 psi (McClintock 1966).
The stress-strain behavior of polymers can be represented by a system of dash-
pots and springs as shown in Figure 7. At low temperatures the dashpots become
more and more frozen, resulting in very little flow. Near the glass transition tem-
perature the flow is established and the material behaves viscoelastically (Fig. 8).
During viscoelastic deformation, the stress-strain relation is strongly dependent
on the strain rate . A high strain rate leaves the polymer little time for plastic flow.
The higher the strain rate, the more elastic and stiff a material becomes. At lower
strain rates dissipative effects connected with viscous flow result in hysteresis of

ELASTIC
F, IT) ~
I
Rubbery/
Flow /

Leathery Rubbery I Liquid VISCO-


// I/ELASTIC
II F2 (TI ~ F2I)e (MEMORY)
(}

-q, ITI

Typicol Amorphous Polymer

I1, TI

200 400 600


Temperature (K) a

Figure 7. Schematic presenta-


Figure 6. Thermomechanical curve of a tion of viscoelastic properties
simple polymer. of polymers.

..................
T-42 T77K
~~~.~ ~ '0 2017

Sec Sec
/ rfrrcture
-
N ' / .

:44 T-293K
i-0 36 -Sec

O 10 12 %
g . O 6
x .'"
S7RAIN E

Figure 8. Stress-strain behavior of epoxy resins at differ-


. ent strain rates, i, and temperatures, T (after Hartwig
1979).

the loading and unloading curves. Figure 8 also shows that at low temperatures
epoxy resins exhibit a linear elastic behavior up to the fracture point; there is no
hysteresis effect and no strain rate dependence (Hartwig 1979).
The behavior of polymers is exceedingly complex. The molecular structures that
influence this behavior are (1) branchinig, (2) chain length (molecular weight), and
(3) crystallinity. The influence of low temperature on these parameters and the re-
sulting change in behavior have been studied extensively in recent years (Hartwig
1979, Frank 1979). Appendix B ,ummarizes the low temperature behavior of a
number of polymers.

, ComTosites
The term "composite" refers to materials having overall properties that are some
average of the properties of several distinct components. One of these components is
contiguous and forms a matrix interfacing with the reinforcing elements. In low
temperature design, composites are particularly attractive because of their low ratio
0' of thermal conductivity to elastic moduluz; oi strength and their high ratio of elastic
modulus to density and strength to density.
Theoretically, both filamentary-reinforced materials, such as fiberglass-epoxy,
and aggregates, such as concrete, fall within this definition. Figure 9 gives a sim-
ple classification of composite systems. Discussion in this section will primarily
relate to polymeric composites.
*The material properties of fiber-reinforced composites depend on (1) matrix type,
(2) fiber type, (3) fiber fraction, (4) fiber orientation, (5) fiber distribution, (6) fiber
I6' .0

,,.,'.',:' ,,.6,'
-'.2 -. , -•. .. --: -'-2 .'.. ': .-
%

.. .. .'.".., ". . ....


. "... ... ",
"Q . .. "..-". .. '. .- . '.
-,'....'.-,
V~~~~~~~~~J ~-Jv~w . Qw-p7- ' W- - 0XrawJ w -J-r~

iCOMPOSITES

reinforced reinforced inforced

nee mareinorceng

products -e.g. rod, fabrication as


pipes, angles, etc. per component

L'" components
This produt will Low-modulus High-modulus
bdesig
, need mchininga reinforement, reinforcement,
e.g. fibers of advanced composites
glass, cotton, etc. e.g. fibers of

Figure 9. Simple classification of composite system.

* strength and (7) temperature. Most of the properties are anisotropic and follow the
general trend of increase in elastic modulus, tensile strength, flexural strength,
and compressive strength with decrease in temperature. The properties of compos-
ites can be controlled by construction techniques, and customized material devel-
opment is possible. Because composites are relatively new structural materials,
their use was initially restricted to selective replacement of specific metal compo-
nents, but this conservative approach is gradually being abandoned.
Low temperature (to the cryogenic level) behavior of composites has been exten-
sively reviewed by Kasen (1975, 1983). He has observed that there is no systematic
data base for lower temperatures. Existing data show extreme variability in
strength properties as a result of embrittlement of matrix resins. Efforts to develop
matrices with improved low temperature toughness continue; however, achieving
this objective remains elusive (Hartwig and Evans 1982). The majority of the cur-
rent data on polymeric composite strength are for room temperature and for the
boiling points of cryogens.
The large variability in composite strength data is due to embrittling of the poly-
mer matrix in cooling. Variability is also a strong function of the quality of lami-
nate manufacture, for example, percentage of void- and resin-rich areas. Quality
control during manufacture can therefore largely control the strength properties of
composites (Kasen 1983).
The anisotropy of the reinforcement results in anisotropy in the properties of the
composite (Fig. 10). Therefore, composite data are customarily related to fiber ori-
entation. Figure 11 shows the temperature dependence of two glass-fabric-rein-
forced epoxy composites in two different orientations (Ledbetter 1979).

nat S. manufcture
-k fo ea e g ., a are P A A 1aresQait
IL.%- - NK U . -MA

,% Fu 1.rntnen 6
- -G I1CR
-- ! CR

I Normal
14 Direrlon

.'.

00 00 200 300
FILL. GPa TEMPERATURE. K

Figure 10. Orientation depend- Figure 11. Temperature de-


ence of composite property. pendence of composite pro-
perty.

FRACTURE PROPERTY TESTS FOR


-e LOW TEMPERATURE APPLICATION OF MATERIALS

Because of economic competition, weight premiums, and the availability of


materials with high yield strength, structural engineering designs are tending to
incorporate higher nominal stresses. However, this trend has also increased the
probability of catastrophic failure. Safe operation of structures at high stresses re-
quires an accurate definition of the fracture resistance of materials to ensure that
an adequate safety margin is maintained in material performance to balance the
effects of design refinements, fabrication, quality and reliability of inspection.
Fracture resistance depends highly on the conditions related to environment,
constraint and strain rate. Fractures can occur at stress levels below yield stress or
above yield stress, depending on flaw size, section thickness and strain rate.
% In order to discuss the effect of temperature on fracture let us recall Figure 4b and
show it schematically in Figure 12. It will be seen from Figure 12 that a material,
when stressed at temperature Ti, will deform first by plastic yielding. If it were to
strain-harden to the curve ao , it would then undergo brittle failure at stress crF. If
the temperature Ti is below TA, the nil ductility temperature (NDT) he material
will simply fail without any plastic strain when the stress reaches aF. We have
already mentioned that BCC materials are far more sensitive to temperature than
FCC materials. Strain rate tends to increase the yield stress ay , displacing the Gy
curve upward as shown by the dashed line in Figure 12. This will lead to brittle
* failure at higher temperature since TA will have increased.

Early fracture test approaches


The foregoing discussion shows that the fracture properties of a material depend
on test conditions (temperature, strain rate) and specimen conditions (flaws).
Early test methods like Charpy impact testing (ASTM 1966, Standard E-23) were
_ therefore designed to introduce a given flaw into a specimen and to test that speci-

%.%
'DT

. Strain-irate

r ef
irate
I Y"e
Str inn
I'a

T2
emp e ra t u re - fracture stress.

2 2

StrainI T2

Temperature

Figure 13. Temperature effect on ductility.

men at a high strain rate. Both the flaw and the high strain rate increase the nil
ductility temperature, and the test therefore gives an NDT that tends to be con-
servative. The energy under the stress-strain curve is used as a measure that can
be related to ductility. A low energy is indicative of low ductility, as shown in
Figure 13a. The energy measured is really the amount of energy used in causing
* plastic flow. Figure 13b shows the difference of energy at two different temperature
levels, T1 and T 2 (T1 < T2 ). The Charpy impact test is a simple and inexpensive test
that is easy to carry out in acceptance testing at the steel works.
The most common specimens employed with the Charpy test in recent years have
been of the V-notch or keyhole type. A specimen is machined from the plate or other
piece of material being tested, cooled to the desired temperature, and placed in the
• testing machine in such a way that it is supported on both ends as a beam (Fig. 14).
A single blow is applied to the middle of the specimen, opposite the notch; as the im-
pactor swings through, the specimen breaks and a recording dial indicates the ab-
sorbed energy. Normally, three specimens are tested at a given temperature and
additional tests are made at selected temperatures. A curve showing the mean val-
* ues of absorbed energy versus temperature is plotted as shown in Figure 13b.

% % %
04

,- %%,%
", , %- ,-,,. , . . % , , % ",% - .. "% -. .. " 71 .% 2. , ,. L, _% L. ,- ,, % '
v~r ~ ~ V~ U W % ~.U.
W~I WW U ~ WVU.W W w WWW T -Z. ' f KX~ ~~~~ ~'~ YW'W

P
,'of
1 I'MPOC

''J

Fiur 14.. Sipeba Ca m


-MM/

" Figure 14. Simple beam Charpy im-


pact machine.

The Charpy test is used primarily to establish acceptance specifications based on


fixed energy levels (say 15 ft-lb of energy at 25°F). It usually shows wide scatter
when specimens are taken from different areas of the same stock, and there are
MD great variations in the transition ranges. Also, the test has not been very conducive
to formulating stress analysis theories (Witzel and Adsit 1969, Harsem 1969, Hall
1969). However, to date it has remained one of the most common methodS of tests to
evaluate impact resistance properties of materials.
Another test that has increased in usage is the Naval Research Laboratory
(NRL) drop weight test (Pellini and Puzak 1963). This test can be carried out with
comparatively simple equipment at various temperatures and produces very little
scattering in the results (Harsem 1969).
. The drop weight test employs special beam specimens in which a crack is created
in the tensile surface. The test specimen is generally about 14 in. long, 3.5 in. wide,
and between 0.5 and 1 in. thick. A weld deposit is placed at the center of the speci-
men and notched to provide the initiation source. The test is conducted by subject-
* ing a series of specimens of a given material to a single impact load at selected
temperatures to determine the maximum temperature at which there is a break (go)
or no-break (no-go) condition. The impact load is provided by a guided free-falling
weight with an energy between 250 and 1200 ft-lb, depending on the yield strength of
the steel to be tested and the size of the specimen. The specimens are not allowed to
deflect more than a few tenths of an inch before they reach a stop.
* ASTM (1966) specification E208 describes in detail the provisions for conducting
and interpreting these tests. The nil ductility transition (NDT) temperature is de-
termined as the maximum temperature at which a standard drop weight specimen
breaks when tested in accordance with the provisions of the drop weight test.
Around 1969, the Naval Research Laboratory developed what was called the dy-
namic tear (DT) method of testing with the aim of overcoming the limitations of the

10
I

r% %
Charpy V(Cv) test and providing information of engineering design significance
in the form of a "stress intensity factor" (Lange and Loss 1969). The test method
was considered successful in that for very brittle material, where linear elastic
fracture mechanics is applicable, direct correlation has been established between
NDT energy and static stress intensity factors, KIc (discussed later).
Many other test methods have been used in practice and research during the last
two decades. A review of all these tests can be found in Hall's (1969) paper on the
evaluation of fracture tests.

Fracture mechanics tests


.In recent years tremendous interest has been generated in fracture toughness
testing based on linear elastic fracture mechanics. The theory of fracture tough-
ness is based on the concepts of Griffith (1920) concerning fracture in brittle mate-
rial. Because of much interest and extensive research, the theory of fracture tough-
ness will be summarized here to clarify later discussion.

Theory of fracture toughness


According to Griffith's theory, when a centrally cracked panel (Fig. 15) is sub-
jected to a uniform axial stress a (psi), the elastic energy loss U (in.-lb) of the plate
due to the growth of a crack of 2a (in.) length is given (Nadai 1950) by

U= 7Ca2 a 2
tiE (1)

where t is the thickness of the plate (in.) and E Young's modulus (psi). IfS is the
surface energy (in.-lb) of the two new surfaces created by the crack, then

S =4yat
=4 (2)

where t is the surface tension of the material (in.-lb/in. 2 ).

St
-0

2'

- -- Figure 15. Centrally cracked panel


under uniform axial stress a.

%1

6W
Equation 1 shows that if a is increased, more stored elastic strain energy will be
available to propagate the crack. Obviously, this crack-driving force is opposed by
an equilibrium resistance of the material. When the rate of ia-rease of the crack-
driving force with respect to crack length equals the rate of increase of the resis-
tance, the crack becomes self-propagating. This type of critical condition occurs
without additional external work only when the elastic strain energy released by a
minute extension of crack length aa is sufficient to develop a new & of crack
length. This unstable or critical condition can be defined as

a(U-S) - 0
c3a

and the stress at which this unstable condition develops is called the critical stress
13c.
'-V By use of the expressions for U and S from eq 1 and 2 and differentiating, the
critical stress ac is given by

''' 1/2
c= L (3)
ira

* Although the above argument is true for brittle materials like glass or ceramics,
the case of metals is different. For metals a much greater energy is required to
plastically deform the grains surrounding the crack tip. However, if a plastic
strain energy term P is added to the expression, the new critical stress for metals
1 can be represented (Orowan 1955, Zinkham et al. 1969) by

ac = [E 2+P)/Ia/ (4)

For metals, the value of P is normally several orders of magnitude larger than the
surface tension of the cracks, so that y can be disregarded.
Irwin (1958) used linear elastic stress analysis theory to show that in an infinite
plate the strain-energy release rate Gc , equivalent to P in eq 4, is related to the
critical stress intensity Kc at the crack tip. Thus disregarding y and substituting
Gc for P in eq 4, we have

2 = 2 (5)

where Gc = critical strain energy release rate (in.-lb/in.)


Kc = critical stress intensity factor (psi/in.)

CFC= cross-section stress at crack onset (psi)


ac = half crack length at crack onset (in.).

Equation 5 is true for an infinite plate. For a plate with finite width W the ex-
". pression for the correction factor a (Irwin 1958) is

a [(W/x a4 tan (nac/W)] (6)

12

0%
According to Irwin (1958), eq 5 for a finite-width panel can be written as
4

EG C =K 2 = rat21 2 (7)

Since from eq 6

=- tan '-

eq 7 can be simplified to give

2=
EG c =KC W tan- . (8)

The ac term, a critical crack length, generally includes a correction for the size
of the plastic zone riy at the tip of the crack, i.e.

ac = a c + rj = ac+ K2 (9)
2w2,
f
where a'c is physical size of crack and ay. tensile yield strength of material.
Note that if a calibration is made of the specimen's compliance with crack
length, the plastic zone is reflected in the compliance measurement in an actual
test, and therefore it is taken into account. If a calibration is not made, the value of
ry is estimated through an iterative process by first assuming ac = a'c. and calcu-
lating the first Kc value from eq 8.
" Irwin (1958) has shown that Gc or Kc decreases if thickness increases because of
transition from a plane stress to a plane strain condition (Fig. 16). In thin speci-

Fatigue
Crock

100% Shear

Thinning
In Thickness
Direction

Piane Stress Condition

Fatique
Crock

No Thinning
SN nFigure 16. Plane strain and plane stress
Pione Strain Condition fractures.
13

!:
" mens plane stress situations are associated with a fracture surface which ap-
proaches almost 100% shear. Thus stresses are not developed in the thickness di-
rection to resist deformation. The stress intensity factor is associated with plane
stress or a mixed mode (normal and shear stress). The material property respon-
sible for developing resistance to deformation is not well defined in this case.
With an increase in thickness the plane strain condition dominates, and the
fracture surface becomes flat, with little or no shear lip. The stress in the thickness
direction is developed depending on its resistance to deformation.
Now, remembering from eq 7 that for plane stress EG, = K, for plane strain we
can write

. .EG, = Kc (10)

where 4±is Poisson's ratio.


We now need to discuss the distinction between KI and K 0 •K is termed the gen-
eral stress concentration factor, and K is a mathematical quantity related to the
stress intensity factor of the first mode (the opening mode, with no consideration of
shear but only of cleavage, i.e. a tensile-stress-related fracture opening). The
ability of the material to resist deformation in the cleavage or first mode in the
presence of the crack is manifested in the fracture toughness parameter KI, or Gjc
* which may be termed a material constant. The important distinction between KI
and Kc is similar to the distinction between stress and strength. Several defini-
tions were proposed for Kjc (Strawley and Brown 1965, Strawley 1969). The latest
operational definition by Strawley (1969) is somewhat arbitrary. This defines KIc
as the stress intensity at which the crack reaches an effective length 2% greater
than at the beginning of the test. This is analogous to definition of yield strength in
terms of a specified amount of plastic strain. Defining K in this manner, we can
rewrite eq 10 as
a. . " EG I,=K,2 (1 - p)(

'S.. In the plane strain (thick specimen) test condition there are two general points
that are relevant in calculating GIc or Kc (Strawley and Brown 1965). First, it is
t., assumed that no stable crack occurs and the initial crack length is used in the
calculation; and second, the plane strain plastic zone correction term is taken to be
one-third of the plane stress term, that is, rly = riy/3 in eq 9, so that
02 K
,;'ac =ac + C
'6,,

* Equation 8 (after Zinkham and Dedrick 1969) can now be completely modified to
show that

EGj:cKI1- 2)= 'c(1- 2)W tan --aC +6 (12)


6WW
0~

14

0%
-'a.
" ,..
-.-.-
., ' ' ' ' ',-., , ,: .---":-- .--.
:.,--:".:..',<, ,-" -.-"., ,Z. .. .. ..,"
1"W WW WV U 'P

where a, = half the original crack ltngth (in.)


k= gross stress at the initiation of slow crack growth (psi)
= Poisson's ratio.

In uniaxial tensile tests of an edge-notched or centrally notched specimen, a


phenomenon called "pop-in" (a rapid decrease of load and then increase) has been
observed. Boyle et al. (1962) have found that the "pop-in" load value coincides with
the plane strain parameter Ktc. Figure 17 shows a typical load-crack displacement
curve indicating the pop-in point.
The stress a1c at pop-in and initial half crack length ac are used to calculate the
Gjc or Kic values. Even though the stable plane strain crack growth is subsequently
arrested by the ability of the material to develop a shear fracture, the Kk value may
be computed from the mixture of the plane strain and plane stress (mixed mode)
type fracture thaL develops as the crack extends to some critical length ac.
To summarize the above discussion, we find that the crack tip stress field is the
driving force for fracture, and the magnitude of this stress field is proportional to
the stress intensity factor K. K being a function of crack size, the applied stress and
structural geometry, it can be computed using structural stress analysis methods.
The resistance to fracture is a material property defined as the fracture toughness
Kc. Fracture occurs when K = Kc. Kc is also a ianction of the extent of plastic strain

Failure
/

0
-J

.rPop-in"

Figure 17. Typical load-crack dis-


Oeflection (in.) placement curve.

"LI 2
0
=* I

K I

I__ Figure 18. Transition -,f ',--#ure


8, Thickmnes toughness.

15

( ' , h% %
- . ". . - ." .
" - - ' -.. ' . -
". - . '.- -. -' --
." . .
".
';
at the crack tip. If the plastic zone is small compared to the specimen dimensions,
K, approaches a minimum constant value defined as the plane strain fracture
toughness, Kic. The relation between K, and KI, is shown in Figure 18 (Tobler and
McHenry 1983).

Fracturetoughness test method


A standard test method, ASTM E399 (ASTM 1981), is used to measure plane
strain fracture toughness, Kic. The test is applicable to fatigue-cracked specimens
having a thickness of 0.063 in. or greater in tension or a three-point bending load.
Load vs displacement across the notch at the specimen edge is recorded graphi-
cally. The load corresponding to a 2% increment of crack extension is established.
The Kic value is calculated from this load by equations discussed earlier. The va-
a-: lidity of the determination of the KIc value depends upon the establishment of a
"sharp crack" condition at the tip of the fatigue crack.

Fracture toughness tests for polymers and composites


In the past, attempts have been made to characterize tlf : failure of nonmetals like
polymers, ceramics, rocks and composites using established fracture mechanics
parameters. However, to date no standard fracture toughness test procedure for
these materials has been established.
The complexity of fracture toughness tests of polymer materials arises primar-
ily because of the large plastic zone at the crack tip and strain rate sensitivity
tiviarshaii and Williams 1973). Most work in this area pertains to room tempera-
_ ture conditions, but low temperature studies on thermosetting epoxies (Kneifel 1979)
and structural thermoplastics (polycarbonate) are available (Kneifel 1979, Martin
and Garberich 1976, Tobler and McHenry 1983). Kk values for plastics are usually
lower than 10 MPa-m 1/2 at low temperatures.
Composites generally fail because of the development of flaws during fabrica-
tion that become enlarged in service. At a microscopic scale this behavior is simi-
lar to that of metal because metal crystals are known to be anisotropic, and random
aggregates of crystals are obviously heterogeneous. Thus, in a fracture mechanics
approach, viewing the composites as statistically homogeneous leads to relations
between variables that are useful in engineering description and prediction of be-
havior. Using this fracture mechanics concept, Corten (1972) has developed a
mathematical model for two-material, bonded composite fracture analysis. Al-
though the concept offers a mathematical tool for the future development of compos-
ite systems, the complexity of the problem has so far prevented this approach from
* becoming popular.

Influence of cyclic load on fracture toughness


We have discussed that the basic material property essential to all fracture me-
chanics consideration is the material's fracture toughness, KI , at the temperature
-. of interest. The Kk value gives an estimate of the critical defect size necessary for
0 fracture with a single application of load. But in real life, few structures are built to
withstand only a single application of load. Consequently, the useful life of a
structure depends upon the rate at which an existing defect will grow to the critical
flaw size as a result of fatigue (and also environmental) crack growth. Therefore,
an additional material property basic to fracture mechanics consideration is the
rate at which a crack will grow under the loading conditions of interest.

16

N...%
V~~~~~~~~~F
. .. V*. - . ... A4,d
Clark and Wessel (1970), using their experimental results, have shown that the
rate o' -rack growth (da/dN)is exponentially related to the crack tip stress inten-
sity per cycle (K), as given by

da - C O AK n
dN

where a = crack length


N = number of loading cycles
n,C0 = empirical constants determined from the test data
K = stress intensity factor per cycle.

Because the crack growth is influenced by cyclic loading, the recommended proce-
dure for fracture toughness measurement includes consideration of fatigue crack-
ing at nominal stress less than 25% of yield stress (Rolfe and Novak 1970).

Strain rate effect on fracture


One of the simplest ways to idealize viscoelastic material behavior under uni-
axial loading at an imposed strain rate dcidt is to express this behavior in the fol-
lowing differential equation:

* 1
E dt+ a
dcr
-T
_ de
o", E dt (Maxwell solid) (13)

N, where E = Young's modulus


cr = stress
t = time
F_ = strain
T, = steady-state tensile viscosity.

-. ' For finite increments of strain, stress and time, Ae, Aa and At, eq 13 can be
rewritten as

E - + A-- (14)
Atl 1E At

Now if time of loading At << 71/E, the term Aa/(rl/E) can be neglected, giving

A0 =E -AE (15)
-. At At

which is a pure elastic loading. In this case brittle failure would result at the tip of a
crack when the local stress exceeds the ideal tensile strength. Alternatively, if rI/E
* is much smaller than At, AalAt can be neglected, giving the equation of Newtonian
viscous flow:

Ac E ". (16)
71 /1E At

17

%
.1%
S T.
. , L 7- T

This equation shows that for small strain rates, there is always enough time for the
elastic strains to relax-that is, the viscous deformation will alter the shape and
orientation of the cracks, making brittle fracture less possible. Also, it follows that
brittle fracture in glassy material (viscoelastic) is governed by the time constant of
loading.
%' Many investigators have examined the effects of strain rate and temperature on
material behavior (Duffy 1979, Klepaczko 1979, Nicholas 1981). For materials with
N strong temperature and strain rate dependence, the fracture toughness usually de-
creases with decreasing temperature and increasing rate of loading. Fracture
toughness should therefore be evaluated in the region where it may show its mini-
mal value and the conditions that reveal the transition from high to low values of
fracture toughness should be investigated.
In order to develop a suitable testing technique for high-strain-rate loading,
Knott (1973) and Klepaczko (1979) have defined the loading rate parameter K1 to
express how fast the crack tip region is loaded:

=
k',I Kltc (17)

where KI, is the crack tip stress intensity factor in mode I (plane strain fracture
toughness) and t, is the time interval from the start of loading to the point when sta-
0 ble crack propagation starts.
The spectrum of rate of loading as defined by Klepaczko falls into the following
categories:

I Quasi-static loading 1
1 MPa -5' 5 3 MPa 1'ss
(with a closed-loop
static dynamic testing
machine)
4 1 * -1
II Instrumented hammer 10 MPa-m-s 5 kK!5 10 MPa #i-s
(for example, instrumented
Charpy hammer)
"- 6 -1 9 -1

III Stress wave loading 10 MPa f-m s < k, "0MPa fms

Obviously, the KI spectrum covers about 10 orders of magnitude, and requires a


number of experimental techniques to be covered. For a K63 Al-Zn-Mg-Cu alloy,
Klepaczko (1979) demonstrated that values of Kit measured at the stress wave load-
". ing are 20 to 36% lower than those for static loading, and for another aluminum al-
Sloy (PA6) this value is almost half of that for static loading. However, Klepaczko's
[ •tests were all done at room temperature. Further reduction in fracture toughness is
expected at lower temperature.

.. 18

%-% %
0%

''=. ""w • " ,.% ".",% % % %' . o % -. % " .. .


". -". . . . . . . . . ." " " '" *"" * " -
TES'I PLAN

Rationale
In the past decade, fracture mechanics principles have been used to quantify
safety factors in structural design, taking into account crack propagation and/or
brittle fracture. Most structural members, components, vessels, and piping are de-
signed according to analysis criteria that guard against failure, but variability in
manufacturing processes of materials may introduce defects or flaws. Despite
nondestructive testing one must realize that flaws do escape detection on occasion
% and may grow to critical sizes under cyclic loading and stress corrosion cracking.
It becomes important, then, to determine the associated minimum critical stress
intensity factors, KI, that will lead to unstable growth in these cracks and also to
know the minimum stress intensity factor, KIa , at which these cracks will be ar-
rested in a particular material.
Klepaczko (1979) has shown a significant difference between the static and dy-
namic determination of Ki, at room temperature; at lower temperatures this differ-
ence will possibly be higher, with Kic showing lower values. In cold regions num-
erous field conditions may be cited in which dynamic loading on structures as-
sumes critical importance. Wave action, wind action, ice sheet movement and
impact, drilling vibration on drill rigs, vehicular motion, blasting and earth-
quakes are a few examples. The hulls of ice-breaking vessels are subjected to re-
* peated impact loading by ice, resulting in deterioration and the need for frequent
repair. Therefore, sound engineering design for cold regions should depend on
material property data determined at low temperature and at high strain rate load-
ing. The CRREL test program aims to generate these data. The next section dis-
cusses this test plan.

Experimental program
The proposed testing system is summarized in Figure 19. For each material se-
lected to be tested, a number of test specimens will be prepared. The number will
depend on the expected variation in test results. Enough specimens will be made so
that statistically meaningful representative data can be obtained from each type of
test.
It is proposed that static, quasi-static and high-strain-rate tests be conducted to
evaluate the fracture toughness property of each material. During its service life at
low temperature a component material is subjected to repeated low-stress-level
loading and occasional high-strain-rate loading. The testing should reflect a
similar loading pattern. Therefore, some specimens will be subjected to initial fa-
.S tigue loading by (tension-tension) load cycling. A comparison of fracture tough-
ness data between the load-cycled and non-load-cycled specimens will reflect the
effect of servi:, life on degradation of material property.

r,Quasi-static fracture toughness test


* The quasi-static fracture toughness test (1 MPa 4i s-1 to 103 MPa 1m s-1) is pro-
posed to be carried out using modified ASTM compact tension (CT) specimens as
described by Klepaczko (1979). In this method, shown schematically in Figure 20, a
compressive force is applied to a wedge-loaded specimen. The vertical displace-
ment of the wedge is measured. Also, the transverse displacement of the specimen
opening can be measured by a clip gauge. The measurements can be recorded by

19
-V

Ma-terial
%saml

Temperature soaking
230 to 270C
0
to - 5C
0
- 200 to - 25 C
0
- 350 to - 40 C

Figure 1.test prora tceat

impt s - ~c~dr

Figurne 20.sur Deiefoausisaicso


aopesinfatlaigmhefrePatnnthed

applieb a tesetinmahie
wedgesis,
Pemnn6aaBs

Mouu

Stent
I

two x-y recorders. in addition, an oscilloscope can record the force-time curve
driving fast crack propagation.
In the configuration of a wedge penetrating the specimen, the tensile force, Fc ,
acting on the crack is given by

Fc (18)
2 tan( . +

where a = half the wedge angle (degrees)


= angle of wedge/specimen friction (degrees)
Pc = compressive force acting on the wedge (lbf).

And according to Klepaczko (1979), the fracture toughness KIc can then be denoted
by

=j Pcfa/W) (psi G 7 (19)


2B W-tan -+ 0
* 2)

where a = crack length (in.)


W = width of the specimen (in.)
B = thickness of the specimen (in.),

.(aIW) is given by
•1/2_ a/2 . 7/29/
a 185.5 + 55.7 - 1017 + 638.9 a
f(a/w) = 29.6

where B > (Kic/Gy)2 and ay denotes the yield limit of the material.

High-strain-rate tests
The high-strain-rate test is proposed to be carried out by using the split Hopkin-
son bar apparatus (Kolsky 1949). In this technique, a small cylindrical specimen is
4 sandwiched between two long bars. A compression pulse generated by impact from
a third bar propagates down one of the bars and through the specimen into the second
bar. The bars remain elastic, although the specimen is deformed into the inelastic
region because of the impedance mismatch. The equations of one-dimensional
elastic wave propagation in long rods along with the recorded strain gauge signals
from the two bars are used to determine the time history of both force and displace-
ment at the ends of the rods contacting the specimens.
In Klepaczko's (1979) stress wave loading technique, the geometry of the test
specimen is similar to that of an ASTM 399 fracture toughness compact tension
(CT) test specimen with the pinholes cut off and an angular incision milled to ac-
commodate a wedge for loading the specimen (Fig. 21). The wedge allows for
4 application of compressive force, whereas the crack tip remains loaded in tension.
The displacement of the wedge, as a rigid body, denoted by UA is given by
21

r -.,.,
- :,
. , 5,,> :
=. € ",'
;._[ -w . :, , . .. , .. ,.; .: ( ,o: . , ..... ..... ,._. ,. ,,,, . ,_.
1
. ~iu
3
flIl W w -vv-PUNSs
V'- J.-~ -- -UWn
;.pY vur -_V i-W-u-- r~rw 'Kr 7' w i' -rWrW pww'r r w

Incident Bar I Tronsmit Bar 2

StrikerShock
Stritker Absorber

SolenoidT
Valve

"% X/At

Ccmpressor N2

"5 Pneumatic Hammer System Nitrogen Cooling System

',# Test Specimen

Figure 21. Split Hopkinson pressure bar applied to fracture dynamics.

UAt) =Co tIt) - ER dt (20)

*and the displacement of the transmitter bar face, which is backing the specimen
(denoted by u]), is

UBt =CO f T(t) dt (21)

where EI , ER and ET are the elastic strains of the incident, reflected and transmitted
pulses, respectively, and CO (in./s) is the elastic strain wave velocity in the bar
given by

Co = (22)

where E denotes Young's modulus (psi), g is the acceleration due to gravity (in./s 2 )
and p is the density of the bar material (lb/in.3 ).
The net displacement of the wedge as a function of time is given oy

=uU(t) (t) - UB (t). (23)

Considering eq 20, 21 and 23, the ret displacement 8(t) (in.) is

8(t) =C O J i(t)- ER(t) dt - ET (t)dt. (24)

For conditions of equilibrium

-I (+ ER(t) = ET(t)-

o2

. . . . .

e; "L " :- : . ' - .? --." .- ": : -: 'i:; : -: -"::::% : "'-i :i ''::-i . ?.4'--- %.:
i.:i l i: ? ~ ~ .: -": i : ., .:y: :: ---
lim W W: WW _ ,
_-M- SS 6 Wb31.1.WL WL L -VW
t._ w ' .... ~ ,p - .W r .r.n. rr.p.-

4i Rewriting,

S 1 W(t)- eT (t) =- R (t). (25)

Substituting the above result in eq 24 above

8(t) = -2C 'eR(t)dt. (26)

If we assume that the bar cross-sectional area is A (in.), PA(t) = the instantaneous
force (Ib) at section A, and PB (t) (Ib) = the instantaneous force at section B, the
forces acting on the specimen are the following:

4, PA(t) =EA i(t) + CR(t)] (27)

and
.
PB (t) =EA [T (t)]. (28)

The average force P(t) on the loading system is

P(t) =- 1.A + PB (0) (29)

From eq 27 and 28
o,

P _t EA [it+ Rt+ Tt].(30)

Again for equilibrium conditions, eq 25 is satisfied, and using this equation in eq


30 we have

SP (t)= EA T (t). (31)


4.'

It will be seen that eq 28 and eq 31 are the same; therefore

SP(t) = PB (t) (32)

Equation 32 shows that the average force is proportional to the transmitted force,
which can now be measured to determine the KI, value. If the critical point (pop-in)
on the ET(t) trace can be detected, the critical force Pc can be calculated:

23
4-,

!0
P (tJ =EAETC(t (33)

where t, is the time period after which the crack starts to propagate. Once P, is de-
termined, Kic can be calculated from

,S{Ki, = EA TC(tcf( a/W) psi. (4


(34)
2B f tan[c2) + tan -1 1.1
where a = wedge angle (degrees)
W = width of specimen (in.)
a = half crack length (in.)
p = coefficient of friction between wedge and material surface
B = thickness of specimen (in.).

Thus, if the coefficient of friction is known, the fracture toughness KIc may be
calculated from the record of P(t) or P(8) where 8 is the wedge displacement.

Test equipment
-3 3
The complete range of testing from quasi-static (10 strain/s) to high (10
* strain/s) strain rate will require loading equipment, specimen clamping devices
in a variety of configurations, transducers, sensors, suitable recording instru-
ments and signal analyzing devices. Table 1 lists this equipment, instrumenta-
tion and facilities.

Table 1. Mechanical equipment and instrumentation require-


ments for low temperature testing of materials.

_'ype of equipment Description

Mechanical 1. Universal testing machine with necessary


accessories, 200 kips (intermittent use-
MTS testing machine available at CRRFL
will suffice)
2. Fracture toughness and fatigue testing
machine, 20 kips (continuous use), includes
load frame, actuator, hydraulic power
supply, load cells, strain transducers,
grips and fixtures, environmental chamber
and control console
3. Charpy testing machine

Instrumentation 1. Four-channel, 500-kHz digital oscilloscope


with signal storing, analyzing, plotting
and computer interface capability
2. High resolution camera
3. High resolution microscope

Fabrication material 1. Testing frame angle irons and miscellaneous


fasteners
2. Impactor/gas system
3. Electronic parts, strain gage and accessories
4. Pressure bars

24

o-0
k.

A large number of tests to determine primary elastic material properties at room


and low temperature will be conducted using test specimens in the MTS universal
' testing machine currently available at CRREL.
A study of the effect of quasi-static strain rate on fracture toughness will be con-
ducted, both with pre- and post-fatigue conditioning of the specimens. Fatigue-test-
ing, tensile testing and fracture toughness testing will be carried out using the 20-
kip machine.
Composites and plastics cannot be tested for brittle fracture/toughness using the
20-kip testing machine. For these tests it is necessary to use a Charpy testing ma-
chine that can be remotely operated in a low temperature environment chamber.
High strain rate studies will be conducted in the Hopkinson bar test setup in-
stalled at CRREL. The specimen will be jacketed for low temperature testing. The
test setup will be designed for use of two 8-ft-long pressure bars, one 1-ft-long impact
bar and an impacting mechanism using a gas gun and associated remote control
valving. The instrumentation consists of the following:
1. Impactor velocity: photo-diode-activated timer.
2. Triggering circuit electronics: accelerometer/vibration pickup.
3. Longitudinal, torsional and flexural strain wave pickup strain gauges.
4. Reaction pad instrumentation: dynamic load cell.
5. Digital oscilloscope to record incident, transmitted, and reflected stress wave
pulses, and to integrate.
6. Associated electronic parts for connection and interfacing of various equip-
ment.

EXPECTED INFORMATION

Expected information from the tests on each material will include its behavior
data, within both elastic and inelastic limits. Within elastic limits, studies will be
made of the change of modulus or stiffness with decreasing temperature and in-
creasing strain rate. Conventional uniaxial tension, compression and flexure
tests and Charpy V-notch test results will establish the baseline material charac-
teristics at decreasing temperatures. For engineering design input, data will be
generated by fracture mechanics tests. Results of these tests will provide a stress
I.-. concentration factor at the crack tips, KIc values, and the minimum crack size that
can be tolerated at various temperatures and loading rates.
Data from fatigue tests to failure at various low temperatures will be generated to
characterize the materials. In order to reflect the conditions of failure of materials
* in service, fracture mechanics tests will be performed on specimens after subject-
ing them to a given amount of cyclic loading
The initial emphasis of the testing will be on polymeric composites; standard test
methods are not available for these materials. Although testing of steel and other
iso tropic materials has achieved high sophistication through fracture mechanics
tests, fracture mechanics of composites are not well understood. Like the material
• itself the art of composite material testing is also in a development phase, and the
proposed test program will contribute to this evolutionary process.
Low temperature property data will be of great value to composite structure
designers. Besides basic stiffness and strength data, several types of ancillary
data may be generated di ring the testing process. For example, acoustic emissions
* from the sample may b- studied during tensile loading of composite beams and re-

25

%
lated to its behavior through the fracture process. With infrared imagery tech-
niques crack nucleation spots may be identified and crack growth may be fol-
lowed. X-ray study may show the filament dislocations developing during testing.
Fractography and scanning electron micrography techniques may be employed to
develop understanding of the fracture processes of composites. Room temperature
data generated by these test methods are available but low temperature data are
sparse. The low temperature data will contribute to the basic understanding of low
temperature behavior.
At later dates, special steels and other materials for cold application will be in-
cluded in the test program. Expected results from the tests of structural metals, e.g.
6 % steel, will be based on the approach that the standard fracture toughness testing is a
complex task in its present form in terms of specimen preparation, instrumenta-
tion and recording of the data. This complexity increases more when the tests are to
be done on material at temperatures below or near their nil ductility temperatures.
As a result, Charpy-V-notch impact testing (CV) is still the most frequently used
test method, despite large scatter in the data. We plan to evaluate simple-geometry
specimens in dynamic tests such as the Hopkinson pressure bar test to obtain room
and low temperature fracture toughness data as a conservative estimate of this
property. The Hopkinson pressure bar method is simple and less time-consuming,
and through proper data processing equipment, results are obtained quickly. Spec-
imens can be cooled to their nil ductility temperatures by a refrigerant inside a
* jacketing around the specimen, and the test is finished in a short time because of
impact loading. It is expected that a large volume of low temperature data can be
" - generated at much less cost by this method than by standard fracture mechanics
tests.
Material specimens for tests will be obtained by contacting the Army research
and materials development communities that have the real need for such materials
data. For example, a need for low temperature composite behavior data has already
*. been identified by the U.S. Army Materials Technology Laboratory, Watertown,
Massachusetts, and it has provided the first batch of composite materials for test-
ing. Results of the study will be published as CRREL reports and will be made
available to primary users (AMC, TRADOC) and other users (TACOM, MER-
ADCOM, ARDC) through OCE (DAEN-ZCM). The results will also have a high po-
a- tential for non-mission-related technology transfer to local, state, and other
federal government agencies as well as to private industry.

SELECTED BIBLIOGRAPHY

Argon, A.S. (1974) Statistical aspects of fracture. In Composite Materials:Fracture


and Fatigue (L.J. Broutman, Ed.). New York: Academic Press, vol.5.
Ashby, M.F. (1972) A first report on deformation-mechanism maps. Acta Metall.,
20: 887-897.
ASTM (1966) ASTM Standard, Philadelphia: American Society for Testing and
Materials, Part 30.
,'N. ASTM (1966) Behavior of materials at cryogenic temperatures. Philadephia:
American Society for Testing and Materials, STP 387.
ASTM (1981) Annual Book of ASTM Standards. Philadephia: American Society
""a for Testing and Materials, E399, part 10, pp. 588-618.

S26

:'0,C
I1IX,

Baldy, M.F. (1976) The development of large-diameter high strength line pipe for
low temperature service. Offshore Technology Conference, Paper no. OTC2667.
Berg, K.Rl. (1983) Composites components--most likely to succeed in vehicles. In
Proceedings, Sixth Conference on Fibrous Composites in Structural Design,
AMMRC, Watertown, Mass., pp. 67-70.
Boyle, R.W., A.M. Sullivan and J.M. Krafft (1962) Welding Journal (N.Y),
Research Supplement, 41:4285-4325.
Clark, W.G. and E.T. Wessel (1970) Application of fracture mechanics technology
to medium-strength steel. In Review of Developments in Plane Strain Fracture
Toughness Testing (W.F. Brown, Ed.). ASTM Special Publication 463, pp.160-190.
Corten, H.T. (1972) Fracture mechanics of composites. In Fracture:An Advanced
%: Treatise, (H. Leibowitz, Ed.). New York: Academic Press, vol. VII.
Couch, W.P., J.E. Gagorcik and T.F. White (1983) Structural evaluation of ad-
vanced composite application to hydrofoil strut/foil system. In Proceedings, Sixth
Conference on Fibrous Composites in StructuralDesign, AMMRC, Watertown, pp.
25-65.
Daziel, D. (1983) High technology test bed: Where we have ben, where we are go-
ing. In Proceedings of the Sixth Conference on Fibrous Composites in Structural
Design, AMMRC, Watertown, Mass., pp. 23-24.
Duffy, J. (1979) Testing techniques and material behavior at high rates of strain.
* ., In Procee ings, Conference on Mechanical Propertiesat High Rates of Strain, (J.
0 Harding, Ed.). The Institute of Physics (London).
•."-Favor, R.J., D.N. Cidieon, H.J. Grover, JE. Hayes and G.lVL McClure (1961) In-
vestigation of fatigue behavior of certain alloys in the temperature range room to
-423'F. WADD Technical Report 61-132, Defense Technical Information Center,
Alexandria, Virginia.
Foye, R.L. (1983) Status of the NASA/DOD composite flight service evaluation pro-
grams. In Proceedings, Sixth Conference on Fibrous Composites in Structural De-
sign, AMMRC, Watertown, Mass.
'e Frank, W.F.X. (1979) Spectroscopic analysis of high polymers for use at low
temperatures. In Nonmetallic Materials and Composites at Low temperature (A.F.
Clark, R.P. Reed and G. Hartwig, Ed.). New York: Plenum Press.
Griffith, A.A. (1920) The phenomena of rupture and flow in solids. Philosophical
Transactions, Royal Soc. (London), Series A, 221: 163-198.
Hall, W.J. (1969) Evaluation of fracture tests and specimen preparation. In Frac-
ture (H. Liebowitz, Ed.). New York: Academic Press, IV:1-44.
Hallet, B. (1983) The breakdown of rock due to freezing: A theoretical model. In
Permafrost: Fourth International Conference, Proceedings, Washington, Na-
tional Academy Press, pp. 433-438.
Hansen, O.). (1960) Materials for low temperature use: Short course on process
'V industry corrosion. In Proceedings, National Association of Corrosion Engineers,
Houston, pp. 273-286.
Harsem, 0. and H. Wintermark (1969) An evaluation of the Charpy impact test. In
Proceedings ASTM Annual Meeting, Symposium on Impact Testing of Metals, At-
lantic City, New Jersey, NTIS No. N71-11585.
Hartwig, G. (1979) Mechanical and electrical low temperature properties of high
polymers. In Nonmetallic Materials and Composites at Low Temperature (A.F.
Clark, R.P. Reed and G. Hartwig, Ed.). New York: Plenum Press.
Hartwig, G. and D. Evans (1982) Fundamentals and Applications of Nonmetallic
• Materials at Low Temperatures. New York: Plenum Press.

Z7

.%
Irwin, Gl. (1958) Fracture. Handbuch der Physik VI (FlUgg Ed.). Springer, pp.
551-590.
Irwin, Gi. (1960) Fracture testing of high strength sheet materials under condi-
tions appropriate for stress analysis. U.S. Naval Research Laboratory, Report 5486.
Kasen, M.B. (1975) Properties of filamentary reinforced composites at cryogenic
temperatures. In Composite Reliability, Philadephia: ASTM STP 580, American
Society for Testing and Materials, pp. 586-611.
Kasen, M.B. (1983) Composites. In Materials at Low Temperatures (R.P. Reed and
A.F. Clark, Ed.). Metals Park, Ohio: American Society for Metals, pp. 413-463.
Klepaczko, J. (1979) Application of the split Hopkinson pressure bar to fracture dy-
r.. ~namics. Inst. Phys. Conf., Ser. 47., pp. 201-214.
, Kneifel, B. (1979) Fracture properties of epoxy resins at low temperatures. In Non-
metallic Materials and Composites at Low Temperatures, New York: Plenum
Press, pp. 123-129.
Knott, J.F. (1973) Fundamentalsof FractureMechanics. London: Butterworth.
Kolsky, H. (1949) An investigation of the mechanical properties of materials at
very high rates of loading. In Proceedings, Physics Society (London), vol. B62, pp.
676-700.
Kreibich, U.T., F. Lohse and I. Schmid (1979) Polymers in low temperature tech-
nology. In Nonmetallic Materials at Low Temperatures (A.F. Clark, R.P. Reed
and G. Hartwig, Ed.). New York: Plenum Press, pp. 1-30.
Lange, E.A. and F.J. Loss (1969) Dynamic tear energy-a practical performance
criterion for fracture resistance. Naval Research Laboratory, Washington, D.C.,
NRL Report 6975.
Ledbetter, H.M. (1979) Dynamic elastic modulus and internal friction in G-10CR
and G-11CR fiberglass-cloth-epoxy laminates. NBSIR 79-1609, Boulder, Colorado:
National Bureau of Standards.
4Marshall, G.P. and J.B. Williams (1973) The correlation of fracture data for
PMMA. Journalof MaterialScience, 8: 138-140.
Martin, G.C. and W.W. Gerberich (1976) Temperature effects on fatigue crack
growth in polycarbonate. Journalof Material Science, 11: 231-238.
Martin, H.L., P.C. Miller, A.G. Ingram and J.E. Campbell (1968) Effects of low
temperatures on the mechanical properties of structural metals. NASA Report SP-
5012(01), Washington, D.C.
McClintock, F.A. and S.A. Ali (1966) Mechanical Behavior of Materials. Reading,
Mass.: Addison-Wesley, pp. 546-561.
McHenry, H.I. (1983) Structural alloys. In Materials at Low Temperatures (A.F.
O Clark, R.P. Reed and G. Hartwig, Ed.). New York: Plenum Press, pp. 371-411.
Nadai, A., (1950) Theory of Flow and Fractureof Solids. New York: McGraw-Hill.
Nicholas, T. (1981) Impact Dynamics (J.A. Zukas et al., Ed.). New York: Wiley, p.
227.
Nichols, R.W. and A. Cowan (1969) Selection of materials and other aspects of de-
sign against brittle fracture in large steel structures. In Fracture:An Advanced
* .Treatise (H. Liebowitz, Ed.). New York: Academic Press, V: 233-284.
Parker, E.R. (1957) Brittle Behavior of Engineering Structures. New York: Wiley,
"4. pp. 253-301.
Pellini, W.S. and P.P. Puzak (1963) Report No. 5920, Naval Research Laboratory,
Washington, D.C.
* Plumer, JRL, Nil Schott, J.& McElman and S-X. Driscoll (1983) Design and
fabrication of FRP truck trailer side rack. In Proceedings of the Sixth Conference

28
;4"

+
L'.
' -"./ J- "."-* -•" -" ..- 'j.- -' - . *,,. . . ,.', .'.'..* . *.- :".. . - - .. -e-Je .% , .. C,fr-. % .... '
on Fibrous Composites in StructuralDesign, AMMRC, Watertown, Mass., pp. 23-
24.
Read, D.T. (1983) Mechanical properties. In Materialsat Low Temperature (Reed
and Clark, Ed.). Metals Park, Ohio: American Society for Metals, pp. 237-267.
Reed, R.P. and R.P. Mikesell (1967) Low temperature mechanical properties of
copper and selected alloys. National Bureau of Standards, Monograph 101.
Rinebolt, J.A. and W.J. Harris, Jr. (1951) Effect of alloying elements on notch
toughness of pearlistic steels. Transactions,ASM, 43: 1175-1214.
Rittenhouse, Jl. and J.B. Singletary (1968) Space MaterialsHandbook. Third
Edition, Lockheed Missiles and Space Co. Technical Report AFML-TR-68-205. Air
Force Systems Command, Wright-Patterson Air Force Base, Ohio.
Rolfe, S.T. and S.R. Novak (1970) Slow bend KI, testing of medium strength high
toughness steel. In Review of Development in Plane Strain FractureToughness
Testing (W.F. Brown, Ed.). ASTM Special Publication 463, pp. 124-159.
Sparks, L.L. (1978) Cryogenic foam insulations: Polyurethane and polystyrene. In
Nonmetallic Materials and Composites at Low Temperatures (A.F. Clark, R.P.
Reed and G. Hartwig, Ed.). New York: Plenum Press, pp. 165-205.
Strawley, J.E. (1969) Plane strain fracture toughness. In Fracture:An Advanced
N Treatise (H. Liebowitz, Ed.). New York: Academic Press, vol. IV.
Strawley, J.E. and W.F. Brown (1965) Fracture toughness testing methods. In
Fracture Toughness Testing and Its Applications, American Society for Testing
and Materials, ASTM Special Technical Publication no. 381, pp. 133-198.
Tetelman, A.S. and A.J. McEvily, Jr. (1967) Fractureof StructuralMaterials.
New York: Wiley, p. 268.
pTitus, J.B. (1967) Effect of low temperature (0 to 65 0F) on the properties of plastics.
Plastics Technical Evaluation Center, Picatinny Arsenal, Dover, New Jersey,
Plastic Report 30.
Tobler, R-L. and H.I. McHenry (1983) i'racture mechanics. In Materials at Low
Temperatures (R.P. Reed and A.F. Clark, Ed.). Metals Park, Ohio: American So-
ciety for Metals, pp. 269-293.
Tomin, M.J. (1983) Fracture mechanics approach to minimize risk of failure in
steels for arctic use. In Proceedings,FractureToughness Evaluation of Steels for
Arctic Marine Use (R. Thomson and C.S. Champion, Ed.). Physical Met. Lab.
Canmet., pp. 12-1-12-15.
Witzell, W.E. and NR. Adsit (1969) Temperature effects on fracture, In Fracture:
An Advanced Treatise (H. Liebowitz, Ed.). New York: Academic Press, IV: 69-
112.
* Zinkham, R.E. and JRF. Dedrick (1969) Fracture behavior of aluminum alloys. In
Fracture:An Advanced Traise (H. Liebowilz, Ed.). New York: Academic Press,
IV: 299-370.

29

!N
* APPENDIX A:~ LOW TEMPERATURE BEHAVIOR
OF SOME STRUCTURAL METALS AND ALLOYS.
800 1 1

Yed600

600
2014 T65lIA 4 4Cu0 8Mq 0 8Sn)
0u""
n
a"O_ 2219T851
2 400
400
6061 T651

Cn 5083 H4321

annealed200
200
50?p,0

1099 4
110010 0

00020 30 00
00020300100 200 300
TEMPERATURE, K TEMPERATURE. K
Figure Al. Yield and ultimate tensile Figure A2. Yield strength of six aluminum
*strengths as a function of temperature for alloys at temperatures between 4 and 300 K
annealed, commercially pure, and solution- (Kaufman et al. 1968) (in McHenry 1983).
hardened aluminum and a precipitation-
hardened alloy. Open symbols are yield
strengths; closed symbols are ultimate tensile
strengths (in Read 1983).

o 2219-T6
x 2219-T6 0 2219-T87
o2014-T6
0 a 2119-T6M
00

-
50 0

0 -

20 20
K 10
S %' K 10.

504

~20 If2

-~ 0

-500 -. 00 -S00 -200 -100 0 00o 400 S00 200 00 0 '00

Temperature. F Temperature. F

Figure A3. Tensile properties of 2000-series Figure A4. Effect of heat treatment on 2219
aluminum alloys in the T6 condition (Martin alunatn alloy (Martin et al. 1968).
et al. 1968).
31
z,

0%
I,-Z
~
- -~ ~ 3 '3-~ '~ - ~Mx
.- ,l.w V1' -. 7 KM L-11KW W:Kf W WY -'-V-"-'~.~- .- ~ --. .w-wv 'WV' -J.u

x A356-T6___
E E 9A612-F

79Q- - T

2'

%Temperawre, F Temperatur. F

*Figure A5. Tensile properties of 7000- Figure A6. Tensile properties of alumi-
series aluminum alloys in the T6 con- num casting alloys (Martin et al. 1.9t)8).
dition (Martin et al. 1968).

- I VIXUR61 '94 10. DIAL*0 @OIMPTIN F. HOMSIE 5751


19-400 PSI; a a -% 20 ~CFO
daD m PLASTIC,
0 S LAS? C. STIRM M GAI.ULAnD USIN*
I ~G~ MW3
&NoI Off TRUE 111TWE PLASTI 1111,111.(

I&A

Figure A7. Fatigue behavior of 1100 aluminum (Favor et al. 1961).


r%%IGELIE cce

K3
- LIM6 V4 a UOW MO OOWT* ?4.
TCWILE STI SUS P09 R* -. S Wil

- --
- - -
4 -

4.4

10- - - - - - -

(Favo et--al--------

- -1.90
4990 PS. M -- PATQ
- L T T.116,1 1111 C OL'O SN
#N w 11 111 11 PATUMNS

6-

- I - - - - - - -O

*8 #~ WE lcs

Figur A9 Faiu falr of 202 almiu (Fvo et al91

I3

I V
FL9X I; '/A iL Ol 1o0 COISI4TION TO,
1T1ENSILI %TIN 44,800 PSi; 41* * OW CON;
OYc l.L S NOT E . C I
5C - i
SI--

PP.

.. o-

a'

00II 0 - - - on

, TEMPERATURE.,-

Figure A10. Fatigue strength of 6061 aluminum


(Favor et al. 1961).

-" 0IOU
1 0 - . 1W
1 CV1 4w O ,g uSL&I i
1- - L ASTaaltasi W]i LCULAT9S
To I M$ C/ & ug gAs PLASIC

FATIGUE LIFE. Cyci

Figure All. Fatigue behavior of 6061 aluminum (Favoret al. 1961)

34

0
PLUAI;V4 04. 60.CONDOION TO. TENS"L
STN SknO 1 11 Rl 1. 1 000 41 1; 010L99
#40 RolloO. 3I

4'.o

It

04 --

-40 -4 -a -4;4 40~ 400"


TEMPERATURE, -F

Figure A12 Fatigue strength of 7075 aluminum


(Favor et al. 1961).

Me 0691
dS" w .t"co$wmFFi LD c

IKATG 10901 WR AWA9 SN

1010N A O W90T9PATCNW
I I-

I SS

F0u_ A3 aiu ehvo f77 lmnu Fvre l 91


%4

so5
toM I

as1040.0

0AIU IFcce
FiueA3 aiu eairo 05auiu Fvre l 91
200 A0E N
STA;NtESS STEEL X A553dleI
2000 - ,Ye'd0
15 8Wd. 300 GRAODE
_vp 304 SOLU TION-ANNEALE 0
- ~~~~cold
rolled 50% 2 AAIN TE

- i~ 100
Tpe 304 aneae
1, X
1000
Tvo 304N 012 w1%Ni

0 50

0 M0 20 300 0 100 200 300


TEMPERATURE, K TEMPERATURE. K

Figure A14. Yield and ultimate tensile Figure A15. Temperature dependence of
strengths as a function of temperature fracture toughness for various nona us-
for zone-purified iron, 9-nickel steel, tenitic steels, showing the beneficial ef-
and carbon steel. Open symbols indicate fects of increased nickel content on
yield strengths; closed symbols indicate transition temperature and subtransi-
ultimate strengths (in Read 1983). tion range toughness (in Tobler and
McHenry 1983).

300 I

AISI 31025

m 200
- ~200 -

1r0

II

00 4 300 10 -

200 300
0100
6TEMPERATURE. K TEMPERATURE. K

Figure A 16. Temperature dependence of Figure A17. Fracture toughness of


fracture toughness for alloys, illustrating 3.5, 5, and 9% N steels at tempera-
characteristicbehavior for three different tures between 4 and 300 K (Tobler
crystal structures (in Tobler and Mc- et al. 1976) (in McHenry 1983).
* Henry 1983).

63

...................... %.-.
SC-1x~
I OTATING *8AU; S/UO. PL.ATE4 WOMMULIEES
TE SIL T4: iIS0O PS11 a -I. 1750 RPM;

ft - -Room TEMP. EC31


- -UNNOTCHEO

0)So - -

FATIGUE LIFE, Cycles

Figure A18. Fatigue behavior of 2800 steel (Favor et al. 1961).

M lop - exuag; &A OI WAs 41A~iNIA~ff I MR,.Ala COOL.


*DoorP ton. All OOL. ~Fo 2 NAR COOL.
11TO 1.00
ITELNSIL91 PSI, X .1 100 CFO;
to- PLASTIC; ... o ELASTIC
CALCULATED USING SoIlI AND ARE
E VESS
IGO- NOT TRUE IN iNS PLASTIC RANGE1.[a

a so-

FATIGUE LIFE, cycles

Figure 19. Fatigue behavior of 2800 steel (Favor et al. 1961).

37
lee6al - - F L ; IAI . GA woo" I60
Is6.
AM COOL, 146o 0 NIL A 00WL F
ISO .. tml AIR CiOOL..
I9SIS:II I7*300
I1. RON OFl GY6L9$ ISOT1L 413
-', * I ~ va ni I

-------

410

'fie-

%---

-. " TEMPERATURE, -F

Figure A20. Fatigue strength of 2800 steel (Favor


et al. 1961).
.5

-U U,8

.'

-U
" ' " ' " " "
-U' " ""d"- " " " " " ": " " ,- ' " """" " ""2 , -.. - ' , ' ' ,
AI-I-WV V -V wV rj

I2

h~6.

40-

00

-~FATIGUE LIFE, CYtlS,


Figure A21. Fatigue behavior of 4340 steel.

FL(I~t~ mamin. NUT

5*c - - - ____ ~
TRET!. T ir S.I I ll £1

a.,moC~

I -J
o Carpenter 20-Cb Ane1eo
z2oo-0tp 0

i2 -Nrenea -'Age.

00--

50 -

230

Kt i0 Kt - 1

1PC

I ___ -_ ___
___ ____ ___ ___ __ ___ ____ ___ ___ ___ ___

0 Ag6aree

U200

2020Kt
1

5. 0 Ag 0adee
0 20 20

350 -c -303 20 -00 0 '-00 -0 300 D

Te0praue 'F20 Tepeatre

Fiur A25.__ Tenil proprtie of HP 9--5 iue 2.Tesl poetisoA 5

stee (Mri 198.ste10rtnel.16) 20 Kl

z %

0' 0' z
2000 1

BAR

304 1
1600

S1200

2000

CL CARBONS1'EEL(Fe
0 8%C) 1

CD 1000
400

Zone purdied IRON qun9 e


NIKE EL(ePered

0 n0
0 100 200 300 102030
TEMPERATURE, K TEMPERATURE. K
Figure A27. Yield and ultimate tensile Figure A28. Tensile and yield strengths of
strengths as a function of temperature for three austenitic strainlesssteels (AISI types
annealed, nitrogen-strengthened, and 304, 310 and 316) at temperatures between 4K
cold-drawn stainless steel 304. Open sym- and 300K (in McHenry, 1983).
bols-yield atengths; closed symbols-ulti mate
tensile strengths (in Read 1983).
400 o Annealed%
I I I
IG Agehardened
Wee
, .00 -

0 8STAINLESS STER 200

300 X0 0

z 4
4KA
~ 300-

6
0oo ~00 2 -.

. .00

0 28
34 0 KI 10
<4004 -
-~0 0

0 0

002 0

0 C?

0 I I 4 I 9
100 t0' 102 1O3 10 10 106 101 C'50 .0 30 0 0 0

CYCLES TO FAILURE Temperature, 'F

Figure A429. Fatigue test results for stain- Figure A30. Tensile properties of l8Ni-8Co-
less steels 304L and 304LNV at 300, 78 and 5Mo maraging steel (Martin et al 1968).
4K (in Read 1983). 4
<K sea - I

-ar10 - - -0 - 0 111 m - 40

ISS11 111-44 PON ft -N S

a Ir

fits

,FK - - FATIGUE

FiueA2-aiu bhvo fAS 04sanessel.Fvre l 91

'a---- 2

wP
II- - - x2
TEMPERATURE, -F TEMPERATURE. *F
-400 -300 -200 -100 0 100 200 300 400 500 -460-400 -300 -200 -'00 0 100 200 300 400 50
*6o IaOs 160

140 140

i120

120 120
,Aged, A'tchd Ailed, CArp, V

cold 0Oa_ Z6. 100 - . -- -


Notched

* W~ -Cold Oco." 60%,


~0 Cold 0,o 26% 1o

W hoooled, Nofthed,

60 6

4040 ------ _ _ _

* A,,oe/.d Cost, Chopy Keyhowe

z', loed 'Aged[

*0 100 200 300 400 So0 0 100 200 300 400 500
TEMPERATURE.*K TEMPERATURE, K

Figure A33. Strength of copper (Reed and FigureA34. Impact energy of copper (Reed
Mikesell 1967). and Mikesell 1967).
TEMPERATURE, -F TEMPERATURE, -F
-460-400 -300 -200 -100 0 100 200 300 400 500 -400 -300 -200 -100 0 100 200 300 400 500
160 32in0-

cold Wo,'ked* 1201 2ko, AIOOLWS Or CLASnoryT

- -- - Arvmi or fl -t

Aged'. OPHfT ~
'20 24 -_

- 100 -20 - - - - - - --

O A,,,,oled

60 12- - -

in0- 0 + C eld D ia 4 %

20

TEMPERATURE, *K TMEAUE*

Figure A35. Hardness of copper (Reed and Figure A36. Modulus of copper (Reed and
Mike sell 1967). 43 Mikesell 1967).

% e
ww w ww w w w w ww..h,~~~-u.r.~ -. . ru s -r rr- r- ,- -- s -,rW.-

TEMPERATURE,*F TEMPERATURE, P
* 40 30 20-100 0 00 200 300 400 500 -400 -300 -200 -100 0 100 200 300 400 500

* 140 '-- -40 1 -

,-~ojo,4 'ord90 t,
705,4.O' A o~led,ChorpyV
12C -20 1 ----- ~

100 -- - - - - - .--. 100 - -.. .

- odWorked.?5%-Nd
V)0 7-So - - - - - --

* A.000l0, Notched

* ~ c 0-7N.- 50,Anele

900

100
200
N6-0

60 040,e9/e [ /0 A/o
Col Wok01%

od., V?
-otc 0

~ ~
and ~ ~ ~ ~ Mieel ~97.(Re ~
ndMksll16) -01
TEMPEATUREF1do eMPERTURE FnWdl

40 k3a0

\ ' -3 An ea ed Cl a- V ,

200 - 20 -
90 /, A heted6 0-I0, old
Cnn

0 100 200 00 000


4rnol 50 10 20 00 40 0

(60320

* I 639inAnEIA ir
- - -6M-D514CSldFDrown

0 0 0 0 0 20 0 0 0 0 0

1EPE2TRE 2KTM4RTUE

FiueA91 adeso0u-0bas Re iueA0.Mdlso uZ bas


and
967. ikeell (Red nd Mkes~l 167)

OS-15 - a'le
APPENDIX B: LOW TEMPERATURE BEHAVIOR
e OF SOME SELECTED POLYAMRS

Tensile Stregt

300 40

300% Modulus

200 1300
1000 200

-50 -25 0 32 17 115 150 212


TESTWhG TEMPERATIME (*F)

Figure Bl. Effect of testing temperature on stress-strainproperties


for natural rubber (Rittenhouse 1968).

4000

100 300% Modulus

-50 -25 0 -32 77 115 150 212


TESTWIG TEMPEATURE ('F)

Figure B2. Effect of testing temperatureon stress-strainproper-


ties for butyl rubber (Rittenhouse and Singletary 1968).

45
00

6F -6 F , 5 C",N
1-.
MI
- 20 C
1"/MIN ,5
6.
-

S x T3 3

":i O 1 I i
I i i ! 11 I

0 OO 200 300 0 O 20 300 0 00 200 300 0 l00 200 300


% ELONGATION % ELONGATION % ELONGATION % ELONGATION

Zn6-+0Cr - .C 6-_20*C 6t-r


, - 10o,'MIN - I 5-,O"/MIN, o/MIN
4 ~ - 4 - 04 4
-. - '.I- -W

r 0' or
U' ' ' ' ' ' o- ' ' ' ' '2

V.i I
0 100
% ELONGATION
200 3YjO 1 OO
% ELONGATION
200 300 0
%
'00
ELONGATION
200 300 0
%
100 200
ELONGATION
300

6.- r - s c T6 6-s- -
200IN5 -20"/ MIN . S
.. . 0 . 2 04,
I 0

2 -3 / '00 02 - , 020C IN2- , 01cI


II
0"
. O I I 0

.. 0 '0 00 300 0 0 200 300 0 100 200 300 0 100 200 0oo
% ELONGATION % ELONGATION % ELONGATION % ELONGATION,
OOP ---

Figure B3. Stress-straincurves of plasticized PVC resin (Titus 1967).


Lowering temperature and increasingrate of loading have the same effect. The curve at -5 0C and at a
Ole loadingrate of2O in. /mi. is the same as that of-200 C at a loading rate of I in. / in. Therefore, a 20-
fold increase in loadingrate is equivalent to a 15 'C reduction in temperatureon the tensile properties
* of this material

* Decreasingtemperatureand increasingthe rate of loading changes the curves from soft to weak, to
soft and tough, to hard and tough, to hard and strong, to hard and brittle.

49'6

'I Io'r ne
, . %.- . - - - - - 4'- 9 . 4 4. 9-
" ,% C_ *-.4
- - 4.- %' '
'.000,000

900.000

S700.000

- 600,000
U

500,000
mooo
.000
.00 400,000
(n

-6.000 YIEL 300.000


0

200,000
zD0
.000 JTIMATE IO(000

1 I 0I
-40 0 40 80 '20 160 -40 0 40 so 120 160
TEMPERATURE !F b TEMPERATURE,.F

12,000

U1 0,000

4 LAa'. 00

rz .000

w2 2
2
T ~ 64000

,.,-0 2,000

0 I 0
-40 0 40 so '20 160 -40 0 40 so 120 460
C TEMP.RATURE.eF 0 TEMPERATURE, F

Figure B4. Tensile and flexural propertiesof vinyl chloride (Saran 909) (Titus 1967).

47

% :" : . %U."
:-. % -
..
..p

%
67000t
TEE
/

TOT

- -0 FEP . ,e v F000
-4000 1 -TL
-69F
50000
on 4000
S-- T3F ?5F
000
, ...
w* ,<m'+ -. , *
-J 73F..
3000 -
* z 000 2?

2000 212r 0 1 o 3

212 F $RI PRCN


1000

0 I I I I I I 1
V0 50 100 150 200 250 300 350 400 450 500 550 600
STRAIN IPER CENT)

Figure B5. Tensile stress vs strain (to fracture) of TEE and FEP (Titus 1967).

I.OW

KIAoo00 TRANSITION POINT

I 00,000

100

-300 -10 -400 -50 0 so P0 IGO 00 0 500


TIwIATUnI, e0

Figure B6. Tensile properties vs temperature of Teflon FEPfilm


(Titus 1967).

48

% %

. +.
.-.-...-.. .,+ .,. + -,+
-. - ,,, - . . ' . _. , + .. .
Table BL Effect of crystallinity on Teflon TFE fluorocarbon resins (Titus 1967).

Property -O
Crystallinity,
-
%
-

50 55 60 65 70 75 80

Tensile
Yield Strength, Kpsi( 1 ) 212 1.20 1.14 1.08 1.04 0.98 0.92 0.83
73 1.90 1.80 1.60 1.40 1.30 1.20 1.00
-40 3.80 3.20 2.80 2.40 2.30 2.10 1.60

Elongation, %(l) 212 - 44 39 33 25 - -


72 58 50 40 32 25 14 6
-40 40 35 30 25 20 10 8

Flexural Modulus ( 2 ) 212 26 33 41 49 57 64 72.


Elasticity, Kpei 73 45 67 WO 110 132 154 176
-88 210 240 270 310 340 370 400

(1) ASTM D638-52T modified for test specimens 3/16" x 1/16"x 7/8" for low-and
high-temperature tests. Crosshead speeds of 0. 02 in./min. (Baldwin) for low
temperatures and 0.05 in./min. (Intron) for high temperatures.
(2) ASTM D790

49
Table B2. Mechaical propmties of FE 24nL

Temperature (0OF)

0 -20 -40
0073
Tensile strengt1h (ps.) 8,320 14,820 16,570 17.990

'2-Yield St rength (ps.) 8.320 14,820 16.570 17,990

*Elongzati~.n ('r) 250 23 15 14

Flexu r I Modulus (p. s. i.) 66.500 336. 300 426. 00v 463.000

Moisture; 0. 19 -0. 23%


Data obtained in accordance with ASTM D) 759.

60-
50 2
40

22
0

~ so. 0 5o0 .00 .50 0 so 100


TEMPERATURE. *C TEMPERATURE. *C

*Figure B7. Maximum Figure B8. Variation of modulus


tensile strength of nylon of rigidity in torsion (Q) for two
11 (Titus 1967). typical grades of nylon 11.

TEMPERATURE. *
-40 -20 0 20 40 60 so 100
14000

12000 - TIEWPCIATU.E,
'C
-000-40 -20 0 20 40 40 0 0

00000 j
. - 00 4 00 0

C 2 4300000 C

400~ '0000

6002400 40001
-P0 0 20 9 *000IS - 0 0 0 *0 20 20

500
000
- - - * 000 - - - --- -- . - - - - - . - .

*-200

100-0.&. A.I .. .A.J 1~ 1j r .t


Table B3.Tensile impact of Zytel nylon resins (Titus 1967).

ZYTEL

PropWtv 101O) W rI7 31(2) 37(2) 42(l)


T ens d e i mpl
a ct . ft 1l) s q in .
Dry: at 73 ° F 82 74 79 103 155
at 0 F 93 80 73 90 105
a -40' F 85 88 68 94 107
Conditioned to 50'% RH at 730 F 156 1(4 121 135 300
at 0°F 113 90 73 96 118
at -40 0 F 79 71 86 88 110

(1) Nvon 66

(2) Nylon 610

24
4 -

20 -

x 1 )
01
2 '2

50

o -4

0-so-- i-40 0 40 so 120 160


TEMPERtATURtE. *F

Figure B11. Effect of temperatureof nylon 6/6 (Zytel 101)


on Izod impact strength at 0.3 percent moiSture content
(Ttus 1967).
22.000

20.000

16,.000

Z16.000
I

12,000

VI

10.000
.- 0 IN MIN.

8.000 -20 IN./ MUN

6.000
.02 IN MIN I/Md/
I I" I I I MilN
'-150 -100 -50 0 so l00 150

TEMPERATURE (C)

Figure B12. Yield stress vs temperaturesfor polysulfone


at various loading rates (Titus 1967).

240

% ~~200
-1,0

160 a

.200% STRAIN
.1.. ~RATE
i MIN.').600
S200% STRAIN RATE
1120 -(MIN -1I1

IL

w ".

.,..-160 -120 -80 -40 0 40 so r20 IGO -160 -120 -0 -40 0 40 W10,:0 100
"'-TEMPERATURE (*C ) TEMPtERATURE. (*C)

'."Figure B13. Percent elongation vs temper- Figure B14. Rupture energy vs temperature
""ature for polysulfone (Titus 1967). The co- for polysulfone (Titus 1967).
! • efficient of thermal expansion is 3.1IxlO-,5 in. /in.
.:. -- F over the temperature range -22 °*to +300°T.

!52

% % %.
F. - -- -- I

CL
:0

-60 -40 -20 0 20 40 60 80 100 t20 140


TEMPERATURE. C

Figure B15. Modulus of rigidity for a typical general


polypropylene ASTM D1043 (Titus 1967).

GEEA PURPOSEL PUPS

I 0 ..
so 0 50 100L ISO 200 250 V, I .-- I
z -SO 0 so 100 ISO 200 2SO

Figure B16. Tensile yield strength Figure B17. Flexural modulus vs


of polypropylenes (Titus 1967). temperature of polypropylenes

(Titus 1967).
10

VSKSTOS ILLEO

4
GEE PURPOSE

-o 0 50 100 ISO 200 250


TEMPERATURE.,

Figure B18. Flexural modulus vs


temperature of polypropylenes
(Titus 1967).
53

-% - - %
600 0

S400

_
1
GEAN
1
I:JERNAL
PURPOSE
4

2 I I !V I
060
GENERAL

dI
"UPOSE

, 4MPACT GENRAL PURPOSEEEAPRPS

IMPPCTCT
600z MAT240

03

~30

so-
2006
a2 - 10 60

40 60 120160 -400 40 120-0 0 4 40 0 40


STEMPERATURE (F) TEMPERATURE (F) TEMPERATURE (F)

10 20
-4O I-

-A_

4, 14

x 21 12

hi GENERAL PURPOSE _j

£IMPACT

4 ~ 4

2 2'
0 0"
-40 0 40 so 120 lEO -40 0 40 60 120 60
TEMPERATURE (F) TEMPERATURE (F)

*% Figure B19. Temperature vs mechanical properties for typical general-purpose and impact
grades of compression-molded polystyrenes (Titus 1967).

54

.00VS 0 1
44

T YRI 767TYRIL ?67

0
40 0 40 eo 120 160 -40 0 40 s0 '20 IS0
TEMPERATiRE *F b TEMPERATURE .F

20.000MAXIMUM FIBER STRESS


7'0'.000 IN FLEXURE .PSI

600.000 TYRIL 76? :00-XPE'S


-ZERLON 150
50i0016.000
ZE0O N1.01 .000

400.000 ZRO S .0
12.000 - TYRIL 767
MODULUS OF ELASTICITY. PSI
300.000

-40 0 40 so '20 060 -40 0 40 11O 120 ISO

C EMPERATURE . F If TEMPERATURE. F

Figure B20. Mechanicalproperties of polystyrene copolymers as a function of tempera-


ture (Tius 1967). Note. Tyrit: ac,-ylonitrile--styrenecopolymer, Zerlon: methyl methacrylate -
8tyrene copolymer.

55

%% %
-- - - _. --.- I/.'. - w~1V1%'' I-
I V- -1 V-I-

10000--

9000- _ _ -

0-400'-, ,5

5000- --

3000-- -

2000-- - -L
iNN
100

0 100 200 30 [K1

'4-. Figure B21. Young's modulus of several epoxy-


resins vs temperature (Hardwig 1979).
'.

DOES NOT RUPTUR I


0 .,/PLASTICIZED VINYL 90 OUROMETER AT TEMPERATURE
[14 LU VE Igor
EE LOW LENIW
DENSITY
4' w POLYETHYLENE o LOWYTYL
00 POYEHYEN
r'.- 1Io EVA % 'I

~~~~0 10
10o
-' EVA -0 0 % 0€'s ,o IS- 100 1 4 t 144Mw
0 4 DENSITY
0 O
4 POLYETHYLENE

TEMPERAATUNE ,OC TEMPEiRATURIEo OF

Figure B22. Stiffness modulus vs temperature of FigureB23. Effect of temperature on Izod im-
polyuethylene, polyethylene copolymers and vi- pact strengths of (Marlex) polyethylene (Titus
nyl (Titus 1967). 1967).

I
40
200-
I001

"o60

0.4

_1' Figure
o4 B24. Modulus of polyethylene as a function
I I0* emperture(Titus 1967).
of

-IM .40 0 4o 0 IGO No0 ago 300


TKUPIPIAT UiiR, *F

Figure B2. Modulus of polyertylene--unedctions


lnlo menitu PEtus 1967). DN-co
4
0 HG DICUP
502

1ik04 99dniypoytyee
DI CUP--i y
lik0.911dnsiy oytyee DUCUP-dicryl:

peroxide; TAC-triallyl cyanurate.

57

- - ,- --- ,- S IN ..
101

Sz
,d09

V( z

01

00-10jO -5 0 10 A
TEPRTUE*

'Il
70000

60000

7000

0-

-0040 -0 0 2 0 0 6 0
VTRi 0930ERTUE

A U0
-

23.C

100 f* 600

0 50-50no
100 20 .0 -2 0 2 40 60 8 10
STRAN, %TEMPRATUE'a

50000

00-DENSITY. a wELONGATION RATE DENSIT

'oo 10%I 200 TY030 1000%/Mi


3000 0 --

40 -0 0 2 0 6 S 0-0 00 290 40 AC 60 0

TEMPERATURE. *C TEMPERATURE, OC

Figure B29. Yield strength and elongations (by Figure B30. Tensile properties (ASTM D-412) vs
ASTM D-412) on compression moldings of poly- temperaturefor polyethylene. (Titus 1967).
ethylenes having densities of 0.915, 0.930 and
0.960 (and melt indexes of 1.8, 2.1 and 0.5, re-
spectively) at strain rates of 100%/mmn. and
1000%/mmn. (Titus 1967).

U .'59
'I,"

00
:-7.1.

. lhi

-WI

I-01
' - 20*C

-p00

Figure B31. Stres-strain curves at various


0 10 20 temperatures of low pressurepolyethylene
%
DEFORMATiCo (Titus 1967).

"N

, E .o 10,000
1:.-9000-
o600 TYPE~
= U Gooo I-
6000

TYPE T "
4,,o/ h°°1,i
So6000 TYPE 6000 TP
STMRTU FTYPE
La TYPE I
*U 4000

2000 2 20
hiJ

-4 . 0 40 s0 120 60 -40 0 40 s0 120 160


TEMPERATURE,*F TEMPERATURE ,'F

"400
~500 500,000
4I3 20 10-, TYP
z 40->40,0
* 400- 3 00,000-
u TYPE Xm
%/ W, 200- - 300,000 TP

[eL
TYPE~ ELW,00T3E1TR.F 2

0 0

I.TEMPERATURE, OF TMEAUE

Figure B32. Strength of comprenion-molded polyethylene at various temperatures (Tius 1967).

I;N
OD
106

-00 -40 0 40 so IO too 200


TEMPERATUI, OC

Figure B33. Modulus vs temperature-unfilledcross-


linked acrylate copolymer DPD-6169 (Titus 1967). DPD-
ethylene ethyl acrylate copolymer; DI CUP-dicumyl
pe-ri de; TAC-triallyl cyanurate.

400

200

@40

10

- 4

of4 .CUYA
0.1 - 1

-00 -g0 0 rIURtUm~uI,


so IGO 10.P WO. 20 200

Figure B34. Modulus of ethylene acrylate copoly-


mer as a function of temperature (Titus 1967).

61

% % 5%J
%0

0 0 3 * 5p off

4 4

4.9. 2 ~~~~LAMINATED
.05
PLEXIGLAS
ENTIGADE DEGEES
TEMPEATU
0UI -02
EMPEAT
AIAE
EDC
MR-DEG
LXGA
EXIGA 0

0 II ---- _ _ A

(a) (b).2.015

Z. =tI

-14

w 3

60 45'0 26-20 1"

00

00
*, 0
400,000
,0

300,0001
01 UJ

- 5 040010,00000 29

18,000
"LOY"3
LUCITE 140
LUCITE" 29
"1
:

-J30,000

20,00
0
0 , 2

TEIMPRATUNE[ - (*F)

Figure B36. Modulus of elasticity of acrylic (Lucite) vs temperature


(ASTM D) 638-52T) (Titus 1967).

24,000 NO LOAD SPEED


22,000I~j 0.05 INCH PER MINUTE:-

22,000

210,000 .I
6,000
4

- 14,000

< 12,000

" g,OO0

-94 -56 -22 14 so as 122 ,56 194


TEMPERATURE INoF

Figure B37. Flexural strength variation with temperature of cast


acrylic sheet (Titus 1967).

63

.. -...
. ..- . ......... .. ... ........ ., .. -
-~

iz 10,000 -4p
40000

000
Z
S6,000- -QO(
-

,4.000 - 0O°
' ?.0000'I I qrI, 1 I t
S -60 -20 0 20 60 100 140 10 220 2,0 -60 -20 0 20 60 100 140 I9O 220 260
TEMPERAtURE. *F TEMPERATURE, 'F

Figure B38. Tensile strength vs temperature of Figure B39. Compression strength vs tempera-
polycarbonate molding material (Titus 1967). ture of polycarbonate (Merlon) (Titus 1967).

001

45 a

15, 30 4S
25U -
20 -. 4

0a
N5

"'"~ . ' 401


20 - • - I

,':o-1 0 SO 12120
w I19
21 I0 ,0 0 so 1 IS
Flre QUIEPOCY.*F TIMIElATURI, ,ot

•Figure B40. Effect of temperature Figure B41. Torsional modulus of polycar-


.. on modulus of elasticity of poly- bonate vs temprature (Titus 1967).
L".
,%,3
carbonate resin (Tius 1967).

%m%
0?0

14 z

12 - 500-

2 - 400

I ~ 300
6 -

I " 200

100
2

0 0L n2=
-100 0 00 200 300 75 60 40 20 0 -20 -40 -60 -00 - 100
TEMPERATURE. IF TEMP RATURE IF

Figure B42. Izod impact strength vs Figure B43. Urethanefoam spring rate vs tern-
temperature of (Merlon) polycarbon- perature (Titus 1967).
ate (ASTM) D-256) (Titus 1967).

40

35-

30 -1

-401 F
T*

20 -'

a6 -

0- + 1651F
4
5-
2
0C . 4 Li I I1J
~ 00 000 000 00 10 20 30 40 50
TEMPERATURE IF STRAIN, PERCENT

Figure B44. Urethane foam yield point vs tern- Figure B45. Compressive stress-
perature (Titus 1967). strain of low density density poly-
ethylene foam at various temper-
atures (Titus 1967).

65
0 0

25
m 1

0
0

'. I I5

75 60 40 20 0 -20 -40 -60 -80 -100


TEMPERATuRE IF

Figure B46. Urethane foam absorbed energy vs


temperature (Titus 1967).

0O5LG/CU FT 0 7 LU/Cu PT

20 20I

ISFTC
LU/C T 0

-@0O P

. STRAIN {INIIN) b ST RAIN (ININ}


40 60 4

20" - 2o
30 /
so, 10 20 10 15

0 0
lf CSTRAIN (IN/INbST AI OW N

i"

0% I

2 3 4 5a 2 0 3 1 4 5 6 7
0
50 -- 8 *
V, Figure B47. Static stress vs strain curves for resilient expanded polystyrene foam (Resilo-Pak)
, at temperature extremes (0.5, 0.75 and 1.3 lb Ifft 3 densities) (Titus 1967).
so
,eSF
TEMPERATURE. K
0 so 100 is@ 206 256 300

2511 2

225 PU -9 60
locoJ

206 (Lfernisaet 0i..197?1)40


01I)
175 -*** 1200

1125
at 0i.111551
(Caaphel
PU(J.)P ~4Uthg"a
125 (Cspe/*i. 52

S8O

250 Fiur B48.)"Sher trngh a afuc


__________________________foms it dnsiyprces,6ndon
pu(.L) A -41978).-

25 200 Figure B48. Young'strmodthus a n-


____F __)_ __-_tion___of temperature for PU and PSm
0i. L* (Carr118 R1 * o with density, proces, and oi-
-N -0 TUUIRTUI.
21 10 0 soetation as parameters (Sparks,7)

TEMPERATURE.
K
TEMPERATURE. K
0 so 106 IS6 206 250 300
260 1 1600
240 ]
PS~i)p.5.5rg/,Jextruded -1100
2206

136 PUti/.IpA54.5k/.J
2. (A'vidses at ci. 1973) 10
160- /PUt//h. A-54.59/rjv m
ca (CAPwphali at ai., 1552) 1
.140-600

120 36
tSoo
I PS.-' '51.3kg/rnomedad
/Arvidaaa at a.1..173) f66 03

320

(Stawart at 81_ 1574) PIJJ4,O 3Ukg/iu


(Car,. 19go)
-41 -406 -360 -261 -106 6 36
TEMPENATUHE.*f

Figure B50. Compressive strength as a function Of temper-


A ature for PU and PS foams with density, process and orien-
tation as parameters (Sparks 1978).

66

U.S GVRNET RNTN FFC: 98-50-5-%22

Das könnte Ihnen auch gefallen