Sponsored By

The Materials Analyst, Part 32: A buyer's guide to analytical services

May 1, 2000

9 Min Read
The Materials Analyst, Part 32:  A buyer's guide to analytical services

This series of articles is designed to help molders understand how a few analytical tools can help diagnose a part failure problem. Michael Sepe is our analyst and author. He is the technical director at Dickten & Masch Mfg., a molder of thermoset and thermoplastic materials in Nashotah, WI. Mike has provided analytical services to material suppliers, molders, and end users for 15-plus years.

When processors and end users encounter a performance problem with a raw material or a product, testing the material or product is typically one of the first items that comes up for discussion. It is an excellent way to collect information that may lead to a permanent solution for the problem at hand. The global economy is pushing manufacturing, assembly, marketing, and distribution activities farther apart geographically. In addition, shorter product development times make it increasingly important that a solution is found quickly when a problem does arise. Identifying the need for testing is the easy part; determining what types of tests are needed is a much more difficult undertaking.

For the uninitiated, the landscape of available services is a bewildering array of abbreviations that stand for even more bewildering terms. And as a survey of the market will quickly reveal, these techniques can be expensive. It is important to spend money wisely. A good approach can save tens or even hundreds of dollars for every dollar spent in the lab. A poor approach can cost thousands and still leave the client wondering what to do next. It only takes one of these latter experiences to turn someone away from going to an analytical laboratory a second time.

One of the things that we are asked to do from time to time is review previous analytical work that left the client wondering what to do next. Typically we find that the problem has been overanalyzed. This is not due to any dishonesty on the part of the laboratory. It arises from three conditions.

First, many analysts are good scientists but have little familiarity with polymers. The specialized knowledge that comes from working in a particular field is something that we often take for granted. Just as a molder may be baffled by talk of DSC, GPC, and FTIR, a Ph.D. chemist who works primarily with nonpolymeric materials may not appreciate the significance of a measurement like molecular weight and may not understand the relationship of this critical parameter to properties.

Second, even if the analyst is polymer savvy, he or she may not have any experience with processing. It is hard to envision all of the things that can go wrong in the molding process unless you have had the hands-on experience with those processes.

Finally, the client and the analyst do not talk enough about the problem before the testing starts. Is the part new or is there a history of successful production? How many cavities does the mold have? Is the problem cavity specific? Was there a recent change in the raw material specification? If an analyst believes the problem has a particular cause, he will follow a certain line of testing to uncover that problem and neglect the proper tests.

There is another impetus that drives the trend toward unproductive testing. It is our current infatuation with standardization. Obviously in an increasingly far-flung organization like our world economy, systems are important. The glue to hold the organization together has to come from somewhere. But too often, standards are written by people who do not understand the nature of the materials with which they are working. In this month’s article, we will highlight a couple of cases of misplaced attention. This will serve as a foundation for an extended discussion on what you as the client should expect from your analytical services facility.

An Unreasonable Request
The first case involves a specification for an aerospace application. Our client was developing an alternative material for the skin of light aircraft. This is obviously a critical application, and a lot of testing was required in order to satisfy the FAA that the material was strong enough, had the correct combination of strength, stiffness, and ductility, and could withstand the elevated temperatures and the fatigue that the plastic material might encounter. The proposed product was a composite based on a thermoset resin, and the reinforcement was a glass fiber.

An extensive amount of testing was required to qualify the material, and over a period of approximately one year the hurdles were overcome one at a time. As our client was filing the final reports and paperwork, an "expert" consulting for the FAA came out of nowhere with a demand for a GPC analysis on the polymer.

Now, for those of you who haven’t been following this series, GPC stands for gel permeation chromatography. Like all chromatography techniques, its purpose is to separate things that appear to be irretrievably mixed together. GPC is used to separate the various sizes of polymer chains in a material. Unlike most classical compounds such as water, salt, and sugar, commercial polymers are not made of molecules with a single molecular weight. Instead, they are conglomerates of long chains, medium chains, and shorter chains that form a molecular weight distribution that looks like a normal distribution curve. When we measure viscosity, we are indirectly measuring the effect of the average molecular weight. When we run a GPC, we get to see all of the components that go into that average.

Whether they know it or not, molecular weight distributions are very important to both the processor and the end user. This is particularly true in the topsy-turvy world of polyolefins where new catalyst systems are making different combinations possible for the first time. Narrower molecular weight distributions allow for greater control over property profiles. Materials can be tailored to emphasize a particular characteristic such as tear strength in a film or melting point in a heat sealing material.

At the same time, changing the molecular weight distribution curve also changes the rheology of the material. Narrower distributions result in materials that do not shear as efficiently. Therefore, for a given melt flow rate a material may require more pressure and a higher melt temperature to fill the same part. Property balances also change. All things being equal, a narrower distribution of molecular weight results in a material that is generally stiffer and stronger but may lack impact resistance.

So if this property is of such value, what was the problem with the request from the FAA consultant? GPC involves getting the polymer into a solution. In order to analyze anything by GPC the polymer must be dissolved in a solvent so that it can be injected into an apparatus called a column. Once it is injected, the solution travels along the column. The largest chains in the polymer sample have little affinity for the column medium and they pass through fastest and are collected first at the other end of the column. The smaller the chain, the more it has a tendency to become adsorbed onto the medium and it takes longer to make it to the end of the column. A detector at the end of the column quantifies the relative concentration of the polymer sample as a function of time. Figure 1 shows schematically how the principle works and Figure 2 shows a result in the raw form of relative concentration vs. time.

The problem lies in the nature of the material being tested. Remember that it was a thermosetting material. By definition, thermoset systems are crosslinked. These materials do not dissolve in anything. Certainly any material that remained uncured could find its way into solution, but this will give a skewed picture of the material because all that will be measured is the material that did not reach the desired state of cure. For this reason GPC is an inappropriate technique for testing the molecular weight distribution of crosslinked materials.

The specification had been written for thermoplastic materials to prevent degraded materials, or materials of inappropriately low molecular weight, from being used in critical applications. The experience of the consultant was with thermoplastic materials. But it was useless in this case. It would be a harmless oversight if the request did not have the force of federal regulation behind it. It took a lot of discussion before the agency dropped the demand for a GPC. Once these things are written into this type of documentation, they take on a life of their own and are very hard to reverse. What is even worse about this case, however, is that our client had already found three laboratories that were willing to perform the work! They too had no experience with crosslinked materials and were not aware that these materials will not dissolve.

A Costly Mistake
Our second case involves a manufacturer of fittings made from acetal, which had begun to fail sporadically in the field. The end user was spending a lot of money in product replacement, and worse yet, was being held responsible by the end user for water damage that occurred when the part failed. Two laboratories had performed extensive testing and had come back with either no conclusions, or a laundry list that read like a troubleshooting guide of everything that could possibly go wrong.

After we reviewed the reports we realized that no one had ever done a direct comparison of the molecular weight of a good and a bad product. In unfilled acetal the easiest method for performing this comparison is the melt flow rate test. Acetals are manufactured according to different flow rates, and the higher the flow rate, the lower the molecular weight. And as we have pointed out numerous times in this column, the lower the molecular weight, the lower the properties. This is especially true of ductility, as measured by either an impact test or elongation at break in a tensile test.

The melt flow test showed that the good product had a melt flow rate of approximately 9g/10 min. The failed parts that we reviewed had a melt flow rate close to 30g/10 min. This did not represent degradation; it indicated the use of a different grade of material from the same product line. It was a simple mistake that cost a great deal of time and money. And it had been missed in a sea of other tests.

Thirty-five years ago Jerry Garcia made a profound statement about good music when he said, "It’s not just about the notes that you play; it’s also about the notes that you don’t play." He could just as easily have made the same statement about analytical tests if he had been so inclined.

Next month we will take a look at the bigger picture of deciding which tests should be performed and which ones should not.

Sign up for the PlasticsToday NewsFeed newsletter.

You May Also Like