A Round-Robin Approach to Seal Strength

Figures 1–3: The three most common seal-strength techniques, such as the “free” or unsupported tail (Figure 1) and the tail supported at 90° by hand (Figure 2) or at 180° by an alignment plate (Figure 3). Adapted, with permission, from F88-05 Standard Test Method for Seal Strength of Flexible Barrier Materials, which is copyrighted by ASTM International (100 Barr Harbor Dr., West Conshocken, PA 19428).
(click image to enlarge)

After conducting hundreds of peel tests, users of the tensile machine testing method probably don’t often think about the group of people who 37 years ago created a standard for flexible-package seal-strength testing, F88. The F02 committee on flexible barrier materials of the American Society for Testing and Materials (ASTM) created the standard to determine the seal strength of flexible barrier materials. Today, it is the cornerstone for physical package testing. Millions of seal-strength measurements are used globally to qualify materials, validate equipment, control processes, and accept shipments during incoming inspection.

Since 1968, ASTM F88 has gone through multiple revisions, reballots, and approvals. (ASTM standards are reapproved every five years to ensure that methods are reviewed and kept up to date with industry use.) The original three-page document that briefly defined static and dynamic procedures has expanded significantly to include its first precision and bias statement, and the recognition of a “bending force” that affects results. This effect makes it important to consider the tail orientation of the sample to be peeled. As F88 states, “A significant difference in results has been shown to be dependent on how the tail is oriented…” In the document, the three most common techniques referenced are a “free” or unsupported tail, the tail supported at 90° by hand, and at 180° by an alignment plate. (See Figures 1 through 3.)

The members of the Sterilization Packaging Manufacturers Council (SPMC) Technical Group are well aware of the conflicting-data issues faced by laboratories using different techniques. When F88 was updated in 2000, members from subcommittees F02.3 on food and consumer packaging and F02.6 on medical device packaging agreed there was a need to replace the old precision-and-bias statement in order to encompass the many modifications in the document. Members of the SPMC Technical Group volunteered to design the protocol, produce the laboratory samples, coordinate the labs, and analyze the data for the precision-and-bias statement. The SPMC Technical Group members included representatives from Amcor Flexibles, Beacon Converters, Oliver Products, Perfecseal, Rollprint Packaging Products, and Tolas Health Care Packaging. Other volunteers included Cryovac, DBI Inc., Distribution Dynamics Labs Inc., Eastman Kodak Co., Edwards Life Sciences, Kraft Foods Inc., Michigan State School of Packaging, and Vinatoru Enterprises.

An Interlaboratory Study (ILS) approach was chosen. (An ILS is used to define repeatability—the amount of variation when samples are tested in a same-lab, same-operator, same-equipment, same-day setting—and reproducibility—the amount of variation found between those labs.) After reviewing the design of the ILS with the F02.3 and F02.6 subcommittees, the SPMC technical group proceeded to create the packets of data collection sheets and instructions for release to participating labs.

Figure 4: The measure of repeatability, r, or reproducibility, R, at 95% means that test results are deemed to be not equivalent if they differ by more than the r or R value for that material or method.
(click image to enlarge)
Figure 5: The grand average of seal-strength changes with each technique of handling the tail of the seal (either unsupported or supported at 90Þ or 180Þ).
(click image to enlarge)
Table I. An outline of the series of tests performed in an interlaboratory study.
(click image to enlarge)

More than 2000 seals were made using a method and technique based on the SPMC 002-96 document for preparation of flat-bar heat-sealed specimens. In order to test the F88 method for differences in technique and data, it was important to limit the amount of variation in the sealing process used to create seals. Lab seals, therefore, were preferred over production seals.

The protocol was designed to comprise three independent round-robin test series for each of the six labs. All testing was based on the use of peelable seals. Each segment of the ILS required that a total of 30 seals be tested for each technique. Test series A evaluated a coated-paper-and-film combination for differences in speed of crosshead travel and for two different techniques of tail peel angles. Series B took an uncoated material with the 1073B Tyvek/film combination and kept the crosshead speed consistent while again varying the control of the tail angle of peel. In this series, both average and maximum values were reported.

An additional reverse-direction test was performed to demonstrate what occurs when less-flexible material is bent back and more-flexible material is supported by the backing plate. In test series C, a sealant material common to the two substrates of different caliper (5-mil foil and 3-mil film) was sealed face-to-face in an effort to measure the effect of a change in flexibility to the measured value.

The equipment used in this study included units from a wide variety of companies, including Dillon, Instron, Lloyd Instruments, MTS Sintech Renew, Test Resources, Thwing Albert, and Vinatoru Enterprises.

In Figures 4 and 5, the measure of repeatability, r, or reproducibility, R, at 95% means that test results are deemed to be not equivalent if they differ by more than the r or R value for that material or method. The 95% means that this statement based on the data calculations will have approximately 95% (0.95) probability of being correct.

The importance of this information is that it clearly demonstrates that a significant difference in results has been shown to be dependent on how the tail is oriented. The grand average of seal strength changes with each technique of handling the tail of the seal—unsupported, or supported at 90° or 180°.

Another factor that affects the measurement is a change in flexibility. In this case, the change involves thickness and material differences between a 3-mil film and a 5-mil foil with the same sealant surface. When tested unsupported, the values of the less-flexible foil seal are lower than the film that is tested unsupported. When supported 180°, the measured value response is two to four times as large, but the stiffer foil material shows higher values than the more-flexible film.

Because of this effort by participating companies, customers and suppliers can understand and take steps to prevent or resolve confusing lab-to-lab data issues. Today, when users specify both tail and material orientation in procedures and reports, they can feel confident that the data will agree with data from other labs.

SPMC is made up of U.S. manufacturers of sterilizable flexible packaging and materials. Member companies include Alcan Packaging, Amcor Flexibles, Beacon Converters Inc., Oliver Medical, Perfecseal—A Bemis Co., Rollprint Packaging Products Inc., Technipaq Inc., and Tolas Health Care Packaging.


No votes yet