1、外文资料翻译： PLASTIC PRODUCT FAILURE DUE TO DESIGN,MATERIAL OR PROCESSING PROBLEMS By Myer Ezrin, Gary Lavigne and John Helwig University of Connecticut, Institute of Materials Science Abstract Several examples are given in which design, processing, or an aspect of the material were primary contributors
2、to failure of plastic products. A common pattern is failure to realize the consequences of seemingly inconsequential practices or decisions. Mold design was a factor in some cases. Material factors and processing were involved in other cases. Frequently design, material and processing are so closely
3、 related that failure cannot be ascribed solely to one of the three (8). 1. Introduction In many cases of failure the cause is at least partly due to failure to know or realize the potential consequences of seemingly safe practices or decisions. In many of the cases cited failure occurs at the manuf
4、acturing stage, either in primary processing, such as injection molding, or in secondary operations. All failures can be traced to the design, the material, or processing, assuming service conditions are not unusually severe. The interdependence of the three main causes of failure is such that often
5、 all are contributors. Material and processing are particularly strongly linked.The material contribution to failure may be in the polymer itself or in an additive. Processing imposes on plastics thermal and mechanical stresses that frequently are the most severe a part will experience in its entire
6、 lifetime. Failure is often due to lack of realization of how severe the stresses in processing are and of the effect on the material. Examples are given of failures due to part design, mold design, material selection and processing. 2. Part Design 2.1 Polypropylene (PP) caps for a packaging applica
7、tion required that the top of the cap be flexed substantially due to direct contact with a round ball at the top of the container. Fracture occurred with some caps from the high flexural load and deformation. The gate was at the center of the top of the cap where stress was greatest in service. The
8、design and material can withstand the service stresses only if the material properties are in control, which was not the case. Inadequate antioxidant and regrind use were the main causes of molecular weight being out of control. This case illustrates a failure to realize how readily certain polymers
9、, in particular PP, degrade during processing and that a small reduction in molecular weight (MW) may be sufficient to cause failure. The design played a part in that the fracture initiation is at the gate which is inherently weak.The effect of processing on the material can be monitored by how much
10、 melt index or melt flow rate increases in processing. Generally an increase of more than 10-20% in most cases may be too much, unless the part experiences very little stress in service. The corresponding decrease in MW may be only about 5%, yet that may be more than the design and the service stres
11、ses will tolerate. Fortunately, melt index is a convenient and sensitive test which takes advantage of the fact that melt viscosity is a function of the 3.4 power of MW above about 20,000 MW ( = KM3.4). Another relatively simple test that provides a measure, in effect, of antioxidant content is oxid
12、ative induction time by differential scanning calorimetry (DSC) (ASTM D3895). This test is particularly applicable to polyolefins (PE, PP). Without adequate antioxidant, PP and PE are very susceptible to oxidative degradation during processing. While this case is cited as an example of part design,
13、it also illustrates how material and processing considerations are also involved. Presented at National Manufacturing Week, Design for Manufacturability of Plastic Parts, March 16, 1999, Chicago. PLASTIC PRODUCT FAILURE DUE TO DESIGN, MATERIAL OR PROCESSING PROBLEMS by Myer Ezrin, Gary Lavigne and J
14、ohn Helwig 2.2 An O ring made of plasticized PVC was in contact with a polycarbonate part in an assembly that required that the PC move freely when the O ring pressure was removed. In service there was sticking, i.e., separation did not occur readily as designed. Plasticizer at the surface transferr
15、ed to the PC, which is not completely impervious to plasticizer. In effect, the plasticizer became an adhesive between PC and PVC. This failure was probably also due in part to the fact that plasticizers are less compatible in PVC under pressure. In this case the effect of plasticizer on PC, an amor
16、phous polymer, was not realized, as well as the pressure effect on compatibility. ABS is also adversely affected by contact with plasticizer from PVC. 2.3 Bottle caps were spray painted for color and scratch resistance. The bottom of the caps were to be bonded to another part of the cap with silicon
17、e adhesive. The adhesive failed to bond to the plastic in some cases. The cause was that some spray paint contacted the bottom surface. Waxy ingredients in the paint, for scratch resistance, interfered with the bond that normally would have been made to the silicone. The design and processing did no
18、t take into account the need to protect the bottom surface while the cap above was spray painted. It should have been realized that even traces of contaminant on a surface can reduce bond strength very strongly. 3. Mold Design 3.1 An ABS injection molded part of a syringe needle holder (4) consisted
19、 of two flats on the inside 180E apart. A metal eyelet and tubing inserted after molding were held in place by stress at the flats. The design called for the flats, which are high stress points, to be 90E removed from the parts two weld lines. In some mold cavities the flats were not located as inte
20、nded, so that the flats were at the weld lines, contributing to failure. 3.2 A hollow ABS injection molded part had a top ring of ABS ultrasonically welded into the inside diameter of the part. Some welds had a protrusion at one point in the circumference, which was thought to be flash from the weld
21、ing. These defects occurred with parts from one of a two cavity mold. Lowering the force of insertion of the top ring during welding did not eliminate all defects. Examination of molded parts for frozen-in stress by immersion in acetic acid (ASTM D1939) showed very little stress. A check for out of
22、roundness showed that bad parts were out of round as much as 0.0025, compared to 0.0005 for good parts. Figure 1 is a cross-sectional view of a welded junction obtained by sanding down a welded unit. The failure is a fracture of the outer wall of the molded part, which occurred only with out of roun
23、d parts. Figure 2 is a sketch of how good and bad parts fit together with the insert. Fracture was due to flash pushing the edge of the part outwards as the ring insert was forced down. In good welds all the flash moved downward inside the part. In this case the human failure was not to check if par
24、ts or the mold cavity were perfectly round. 4. Material 4.1 A glass-filled PBT (polybutylene terephthalate) part had a hole in the center in which a threaded metal part moved freely back and forth. In oven aging at 160EC to simulate under the hood automotive service the metal part lost its ability t
25、o move freely in the PBT part, which had shrunken slightly. Shrinkage was due mainly to further crystallization in service beyond the degree of crystallinity as molded. DSC showed that the heat of fusion increased approximately 20%, corresponding to a like increase in degree of crystallinity. The cr
26、ystallinity developed on aging at 160EC is seen as a new peak at approximately 200EC. Shrinkage would not occur if the part was fully crystallized. It would not be a problem if the fit or tolerance between metal and plastic was not so tight. Possibly a nucleating agent in the PBT would give complete
27、 crystallization as molded, so that shrinkage as molded would not occur in service. What was not realized was that crystalline polymers may shrink in service if not fully crystallized. 4.2 A prototype part was machined from a block of plastic believed to be acetal homopolymer. It performed in trial
28、runs in service below expectations. Consideration was being given to redesign or to a change in PLASTIC PRODUCT FAILURE DUE TO DESIGN, MATERIAL OR PROCESSING PROBLEMS by Myer Ezrin, Gary Lavigne and John Helwig material. A check of the material by infrared spectroscopy and DSC showed that it was HDP
29、E, not acetal. The trial run results were consistent with what would be expected of HDPE. The failure was in assuming incorrectly what the type of material was. 5. Processing As indicated in the Introduction, a common failure is not to realize that the most severe and potentially damaging stage in a
30、 plastics entire experience is the thermal and mechanical stresses of processing. This problem is particularly serious for condensation polymers (nylon, PET, PC, PUR) and for polyolefins, although it is a problem for all materials. In the former case, hydrolysis to lower MW can take place if water c
31、ontent is above about 0.01%. The requirement of practically complete dryness in the melt cannot be overemphasized. For polyolefins like PE and PP, oxygen is the enemy, together with free radicals (reactive carbon atoms lacking one hydrogen atom) (5). Without adequate and effective antioxidant, the stage for failure is set in the molding machine or extruder. Section 2.1 above refers to the 3.4 power
