A DIVISION OF
THE SOCIETY OF THE PLASTICS INDUSTRY, INC. • 1801 K STREET, N.W., SUITE 600K, WASHINGTON, DC 20006 • (202) 974-5200
Technical Bulletin
1978
PBI 11
1991
Rev. 1
TEST METHOD FOR DETERMINING TOP LOAD
STRESS CRACK RESISTANCE OF BLOW MOLDED POLYOLEFIN BOTTLES
1.0 SCOPE 4.0 SUMMARY OF METHOD
1.1 This method determines the stress crack The test consists of exposing the finished, heat-
resistance of blow molded polyolefin bottles. The sealed, blow molded bottles to the action of the
test measures the effects of material, blow test solution contained at nominal fill point within
molding conditions, bottle design, contents and the bottles at an elevated temperature and a
other factors that govern stress crack resistance. constant top load.
The bottles are tested under constant top load
conditions at an elevated temperature. The 5.0 SIGNIFICANCE
amount of top load required depends on the
bottle size, weight and design. 5.1 Stress cracking of blow molded containers is
influenced by excessive top loading in storage,
1.2 The purpose of this test is to compare one group surface active agents, internal vapor pressure,
of bottles to another of the same design. physical or chemical attack, molded-in stress, etc.
The results of this test should be clearly
2.0 REFERENCED DOCUMENTS understood to be representative only for the
particular test performed under the conditions
ASTM E 145 "Specifications for outlined. Acceptability of performance shall be
Gravity-Convection and Forced Air Ovens." agreed upon by purchaser and supplier. The
underlying purpose of the Recommended Practice
3.0 DEFINITION is to obtain an eventual correlation with warehouse
storage.
3.1 Stress cracking of a blow molded container shall
have occurred when any of the following are 6.0 APPARATUS-See Appendix 2 Apparatus & Materials
observed: List
3.1.1 It has lost pressure through any opening 6.1 An environment capable of maintaining a
other tan heat-seal areas. temperature of 50° (122°F) + 1°C (+ 2°F)
3.1.2 There is visible, to an observer with normal 6.2 A balance accurate to within +/- 0.1 gram for
eyesight, any crack completely through the bottle weighing bottles and contents.
walls resulting in the leakage of product.
6.3 A multi-station testing device designed to test
3.1.3 There is evidence of the contained product a set of at least 15 bottles, as outlined in Appendix
on the outside of the bottle coming from an 2.
opening other than heat-seal areas.
NOTE 1:Since the lateral bearings prevent
frictional drag from misalignment, they are
To the best of our knowledge the information contained herein is accurate. However, The Society of the Plastics Industry, Inc. assumes no liability
whatsoever for the accuracy or completeness of the information contained herein. Final determination of the suitability of any information or material for
the use contemplated, the manner of use and whether there is any infringement of patents is the sole responsibility of the user.
This test method or practice may involve hazardous materials, operations, and equipment. This test method does not purport to address all of the safety
measures associated with its use. It is the responsibility of the user of this method to establish appropriate safety and health practices and determine the
applicability of any federal, state, and local regulatory limitations prior to use.
COPYRIGHT 1991 by THE SOCIETY OF THE PLASTICS INDUSTRY, INC.
PBI 11Page 2 of 13 Pages
the critical section of the testing device. These bearings outside of the bottle since this might lead to a premature
must be of the type specified or superior quality to insure failure.
consistent and proper loadings throughout the apparatus.
10.2 Heat-seal the bottles with a suitable laminate
6.4 Heat-sealing unit for sealing the laminate to the bottle. and be sure that the plastic-coated side of the
heatseal foil is in contact with the sealing
7.0 REAGENTS AND MATERIALS-See Appendix 2 surface of the bottle. Test the sealed bottles for
Apparatus & Materials List leaks by inverting and applying firm pressure
without grossly deforming the bottle. Cap with
7.1 Heat-seal laminate for sealing bottles. a polyethylene or suitable lined closure with
sufficient torque to maintain the seal.
7.2 Properly seating closures that will hold heat-seal
laminate in place without cutting through its layers. NOTE 5: The pressure applied during
heat-sealing should be a minimum to avoid de-
7.3 A surface-active agent such as Igepal CO-630, a forming the neck finish of the bottle. Any other
polyoxyethylated nonylphenol. deformation of the bottle during sealing also
may affect the final test result.
NOTE 2: Polyoxyethylated nonylphenol is hygroscopic
and the undiluted reagant should be kept tightly 10.3 Condition the filled and sealedM at 50°C
stoppered. (122°F) +/- 1C (+/- 2°F) for 4 hours Place the
conditioned bottles in vented plastic bags and
7.4 Distilled or demineralized water for preparing the test put each bottle in the testing device with a top
solution of Igepal CO-630, a 10% aqueous solution by load applied according to Note 5.
volume of the surface-active agent.
NOTE 6: It is desirable to have the F50 value
NOTE 3: Mixing and storage should be done at room fall into a range of 3 to 5 days for at least one
temperature as excessive heating to facilitate of the groups of bottles being compared.
preparation of fresh solution can lead to hydrolysis. The Increasing the top load accelerates the
test solution should be stored in a closed container to tendency for bottles to undergo ESCR-type
prevent contamination and evaporation. failures. Therefore, the top load required for
obtaining desirable F50 values must be
7.5 Polyethylene bags large enough to completely enclose adjusted for each type of bottle. This can be
a test bottle without exerting pressure on any part of the done by pretesting three levels of top load.
bottle.
One method of pretesting is to estimate the
7.6 Log Probability paper-KE 468082. load that could give the desired F50 value. This
estimate can only be done by guessing or prior
8.0 TEST SPECIMENS experience as there is no method of arriving at
this value from other information. Top loads are
8.1 The blow molded bottles are the test specimens then selected above and below this value to
and should come from normal production and be assist the final selection of a top load level.
representative of the lot to be tested. Polyolefin Fifteen sample units are top loaded with five
bottles of any size, shape or weight may be tested. samples at each load level distributed
throughout the testing device. The test would
8.2 At least 15 randomly selected bottles are required then be performed at the load near the 3 to 5
for each test. However, extra bottles should be day range.
selected for use in screening tests and substitution
of bottles with defective heat-seals. (Mold cavity 10.4 The temperature of the oven should be
of each bottle selected should be noted since they checked periodically. The temperature shown
may have an effect on the results). on the recorder graph or other reading devices
should be verified by measurements with an
9.0 CONDITIONING accurate thermometer at different locations
The empty test bottles and the test solution are inside the oven.
conditioned for a minimum of 72 hours at 23°C 10.5
(73.4°F) +/- °2C (+/- 3°F) 10.5 Based on the procedures described in Note 6,
apply the most suitable top load to a sample of
10.0 PROCEDURE fifteen bottles.
10.1Fill a minimum of 15 containers to nominal capacity with
fresh test solution that has not been used previously. 10.6 Inspect them without handling the bottles for
stress cracking after 4 and 8 hours. Thereafter,
NOTE 4: Avoid spilling the test solution on the inspect at least once every 24 hours, but more
frequently if desired. Bottles which have stress
PBI 11Page 3 of 13 Pages
cracked are to be removed and the time and plotted. The F-50 value lies at the point where the best line
type of stress cracking must be recorded. Test intersects the 50% line on the probability scale.
is to be discontinued after 14 days regardless
of the number of stress crack free samples 11.4 A sample calculation is shown in Appendix 1.
which may remain.
12.0 REPORT
NOTE 7: Inspection time may be minimized by The report shall include the following:
equiping each station with a switch that would
detect a failure. This would be connected to a 12.1 Complete identification and description of bottles
timer, logger or computer to record the failure tested and the top load weight used.
time.
12.2 The F50 value from the probability plot.
11.0 CALCULATIONS
12.3 Location of stress crack failure according to the
11.1 Calculate the F50 value, the probable time at terminology shown in Appendix 1-Figure 2.
which 50% of the bottles have stress cracked.
For the full top load ESCR test using 15 12.4 Any points in which the test was not carried out
samples, the F50 value is to be determined according to the procedure outlined in this
from a plot on log/ probability paper (see 7.6). Recommended Practice.
Plot the time at which the first stress crack
occurred versus the plotting position (%), for 13.0 PRECISION AND BIAS
one stress crack according to the formula:
Attempts to develop a precision and bias statement for this
11.2 Plotting Position (%) _ method have not been successful, because not enough
laboratories were found to carry out a valid round-robin. For
(N;) Number of specimens which stress cracked this reason, data on the precision and bias cannot be given.
(Number of specimens tested) + 1 Anyone wishing to participate in the development of
precision and bias data should contact the chairman of the
11.3 Plot the time at which the second stress crack technical committee.
had occurred versus the plotting position for
two stress cracked specimens and continue in
this matter until all stress cracks have been
plotted. Finally, draw the best straight line
through the points
PBI 11Page 4 of 13 Pages
APPENDIX 1
GRAPHICAL METHOD OF DETERMINING F-50
Stress crack data is plotted on logarithmic probability Occasionally, a specimen may be irretrievably lost. The
graph paper. Elapse time is entered on the logarithmic plotting position on the probability scale will thus change
scale. The number of stress cracks divided by one more but the instructions remain the same.
than the number of original specimens is recorded on the
probability scale. Adding one to the number of specimens
is a widely accepted method for handling data involving
ten or more sample specimens. The technique leads to
symmetrical data treatment, provides for plotting of all data
points and is adaptable to any number of specimens.
If one or more specimens broke since the previous
observation, plot a point for each of the specimens. (An
alternative approach includes the use of timers attached to
microswitches to indicate the exact time of failure).
Therefore, with either approach, one point will appear for
each specimen that stress cracks. A point does not
necessarily appear for each observation time. Draw the
best-fitting straight line for the plot. The time indicated at
the intersection of the data line and the 50% probability
line shall be the 50% point, (F50)
Specific examples of calculations using the graphical
method in conjunction with Figure 1 are given in Table 1.
Avoid overextrapolation when using this graphical method.
In case of unreasonably extended life of some specimens,
the test may be terminated before all specimens have
stress cracked. The available data may then be plotted.
Reduced precision may result.
PBI 11Page 5 of 13 Pages
PBI 11Page 6 of 13 Pages
PBI 11Page 7 of 13 Pages
PBI 11Page 8 of 13 Pages
APPENDIX 2
MATERIALS AND APPARATUS LIST
TEST METHOD FOR DETERMINING TOP
LOAD STRESS CRACK RESISTANCE OF BLOW MOLDED
POLYOLEFIN BOTTLES
A. Fifteen-station testing device designed to the attached
drawings.
E. Weights for use with top load
B. Heat sealing laminate for sealing bottles:
Several systems have successfully used such as:
Berlin Packaging Solid lead weights in 1 or 2 lb. Increments that can be
111 North Canal St. stacked to the desired level.
Suite 300
Chicago, IL 60606 Other material can be used such as steel (stainless
preferred) or brass weights.
C. Heat sealing unit such as:
Clamco Model 50T, 165W:
11350 Brookpark Road
Cleveland, Ohio 44130 Also attached to this M & A Section are:
D. Stress Cracking Agent such as: Figure 1-Bottle Tester Assembly
Figure 2-Bottom Support
Igepal CO-630 Figure 3-Upper Test Plate
Rhone-Poulenc Figure 4-Bottom Test Plate
P.O. Box 643 Figure 5-Stud, Angle Brace, Contact Block, Bill of Materials
Spartanburg, SC 29304-2643
PBI 11Page 9 of 13 Pages
PBI 11Page 10 of 13 Pages
PBI 11Page 11 of 13 Pages
PBI 11Page 12 of 13 Pages
PBI 11Page 13 of 13 Pages
