Initiation of stress-corrosion cracking in unidirectional glass/polymer composite materials

https://doi.org/10.1016/S0266-3538(00)00208-6Get rights and content

Abstract

The purpose of this research is to determine the resistance to stress-corrosion cracking (SCC) of three unidirectional (pultruded) E-glass/polymer composites based on modified polyester, epoxy and vinyl ester resins. The composites have been subjected to a nitric acid solution of pH 1.2 in a newly designed four-point bend fixture. The stress-corrosion process was initiated on the as-supplied surfaces of the composites. The process has been monitored for acoustic emissions and the stress-corrosion surface damage in the specimens was investigated by the use of scanning electron microscopy. Experimental results indicate that the stress-corrosion cracks originate predominantly from exposed glass fibers on the surfaces of the composites. The process can be successfully monitored by means of acoustic emission equipment. Three stages of SCC, crack initiation, sub-critical crack extension and stable crack propagation, can be distinguished by carefully examining the curves of acoustic emission (number of events) versus time. For the first and second stages of SCC, the acoustic emission outputs are linear functions of time. The slopes of the first and second stages of the curves of acoustic emission versus time have been used to determine quantitatively both the resistance of the composites to crack initiation and sub-critical crack extension, respectively. It has been shown in this research that the resistance to the initiation of SCC in nitric acid of the E-glass/vinyl ester composite is approximately 10 times greater than the E-glass/epoxy composite. Furthermore, the E-glass/epoxy system exhibits approximately 5 times higher resistance to the initiation of SCC than the E-glass/modified polyester system. The sub-critical crack extension process is also significantly more rapid in the E-glass/modified polyester than in the E-glass/epoxy composite.

Introduction

Composite suspension insulators are commonly used in overhead transmission lines with line voltages in the range of 69–735 kV [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28]. The insulators rely on pultruded glass-fiber/polymer-matrix rods as the principle load-bearing components. There are two metal end-fittings attached to the rod and the surface of the rod is covered with a rubber housing material with multiple weathersheds. One of the major problems related to the application of composite insulators on overhead transmission lines is their mechanical failures in-service by brittle fracture. Brittle fracture of composite insulators results from stress-corrosion cracking (SCC) of the composite rod material [1], [2], [4], [5], [6], [8], [9], [10], [14], [15], [17], [19], [21], [22], [24], [25], [26], [27], [28]. SCC in an E-glass/polymer composite is a consequence of chemical attack on the fibers by an acid (either organic or inorganic) in conjunction with low mechanical tensile stresses applied along the rod axis [29], [30], [31], [32], [33], [34], [35], [36], [37], [38]. Recently, it has been shown [24] that the brittle-fracture process in composite insulators can be caused by the in-service formation of nitric acid solutions as a result of the presence of moisture and corona discharges. If the acid penetrates the end-fitting and reaches the surface of the composite rod, a brittle-fracture process can be initiated, leading to insulator failure in service. In Fig. 1, an insulator which failed in service by brittle fracture is shown. The most typical feature of the brittle fracture failures is the presence of large planar fracture surfaces in the rods which run perpendicular to the glass fibers.

The characteristics of brittle fracture failures of composite insulators have been extensively studied over the years by Kumosa et al. [14], [15], [19], [21], [22], [24], [25], [26], [27], [28]. Numerous testing techniques and numerical models have been designed to explain the brittle fracture process in various designs of composite insulators. Several different stress corrosion methods have been used to duplicate the process under laboratory conditions. Various pultruded E-glass and ECR-glass /polymer composites have been subjected to the combined effect of mechanical, chemical and electrical stresses. In particular, the effect of either organic (oxalic) or inorganic acids (nitric) on the resistance to the stress corrosion process in the glass/polymer composites has been extensively investigated [28], [37], [38]. However, none of the previous testing techniques used in the insulator research provided meaningful conclusions regarding the resistance of the composites to the initiation of SCC. Therefore, ranking of composites based on E-glass fibers for their resistance to brittle fracture could not be accurately established. Since a vast majority of suspension composite insulators are based on E-glass fiber composites with either polyester, epoxy and vinyl ester resins, the issue of initiation of SCC in these composites when subjected to mechanical tensile loads in the presence of a nitric acid solution (the most probable cause of brittle fracture failures) is extremely important.

The previous tests have shown that the stress-corrosion process can be initiated in a composite subjected to nitric acid and mechanical stresses [28], [37], [38]. Moreover, the process has also been initiated in some of the composites when subjected to the combined effect of mechanical stresses, corona discharges and moisture. Regarding the effect of different glass fibers (ECR vs. E-glass) and polymer resins (modified polyester, polyester, polyester with clay, polyester with calcium carbonate, epoxy, vinyl ester and modified vinyl ester) on the stress-corrosion process, the following two major conclusions were drawn [28], [37], [38]:

  • 1.

    Neither crack initiation nor propagation could be achieved in the ECR-glass/polymer composites under the influence of a nitric acid solution (pH 1.2) and mechanical stresses.

  • 2.

    For the E-glass-fiber/polymer composites, the type of polymer resin does not significantly affect the stress-corrosion process under the stable crack propagation conditions. Stress-corrosion cracks in the composites grow almost at the same rate when subjected to mechanical tensile stresses and nitric acid and their crack growth rates can be described using a power law [31], [34], [35]. Below the stable crack growth regime, the SCC process is still possible, however, the crack propagation process cannot be described by the power law relation.

Despite the fact that differences have been noticed regarding the initiation of SCC in the E-glass/polymer composites depending on the type of resin, the exact relationship between the onset of stress corrosion cracking and the type of resin could not be quantitatively established. Therefore, an attempt was made in this research to evaluate the resistance to SCC in nitric acid, and in particular, the resistance to the initiation of brittle fracture in the three most commonly used pultruded composites in composite suspension insulators. These composites are based on E-glass fibers with either modified polyester, epoxy or vinyl ester resins. The resistance to SCC in nitric acid of the composites was investigated by exposing the as supplied surfaces of the composites to the acid under four-point bend testing conditions.

Section snippets

Fixture design

The stress-corrosion experiments on unidirectional E-glass/polymer composites were performed by using a newly designed four-point bend fixture. A schematic illustration of the fixture is shown in Fig. 2. The load was applied to the specimen by placing weights on the top plate using the pin to ensure they are completely centered. Both the spacing between the load pins and the diameter of the load pins were in accordance with ASTM specification D790-86 standard [39]. Bearings were inserted in the

Stress-corrosion test results and discussion

Three E-glass/polymer systems, namely E-glass/epoxy, E-glass/modified polyester, and E-glass/vinyl ester were tested for their resistance to the SCC in nitric acid.

Discussion

It has been shown in this research that the acoustic emission results are significantly different for each composite tested in nitric acid. The highest rates of acoustic emission signals were observed for the E-glass/modified polyester composite, with the lowest for the E-glass/vinyl ester composite material. There is a strong relation between the amount of stress-corrosion surface damage observed on the surfaces of the specimens and the acoustic emission data. The largest amount of surface

Conclusions

  • 1.

    A series of stress-corrosion tests have been performed on three different composite systems based on E-glass-fibers with modified polyester, epoxy and vinyl-ester polymer resins in order to evaluate the resistance of the composites to stress-corrosion cracking in nitric acid. Particular emphasis has been placed in this research to determine the resistance to the initiation of stress-corrosion cracks by subjecting the as supplied surfaces of the composites to nitric acid in the presence of

Acknowledgements

This research has been supported by the Western Area Power Administration, Pacific Gas and Electric Company and Glasforms, Inc. The authors are especially grateful to Mr. O. Perkins of WAPA, Mr. D. Shaffner of PG&E and Mr. T. McQuarrie of Glasforms for their continuous support of this study.

References (40)

  • Chandler H, Reynders J. Electro-chemical damage to composite insulator. CIGRE 1984 Session, Paper...
  • S.J. Harris et al.

    Metallographic investigation of the damage caused to GRP by the combined action of electrical, mechanical, and chemical environments

    J. Material Science

    (1984)
  • Akhtar A, Nadeau JS, Wang JY, Romily DP, Taggart C. Brittle fracture of non-ceramic insulators. Report for the Canadian...
  • A. Akhtar et al.

    Failure analysis of brittle fracture in nonceramic insulators

    J. Composite Technology and Research

    (1987)
  • De. Tourreil C, Bourdon P, Lanteigne J, Nguyen-Duy P. Mechanical evaluation of non-ceramic insulators. Report no. CEA...
  • Paris L, Pargamin L, Dumora D, Parraud R. Rating of composite suspension insulators related to the long term mechanical...
  • De Decker D, Lumb C. Mechanical strength of composite suspension insulators. In: proceedings of the SEE International...
  • Kumosa et al. M, Micro-fracture mechanisms in glass/polymer insulator materials under combined effects of electrical,...
  • Kumosa M, Qiu Q, Bennett E, Ek C, McQuarrie T.S, Braun J.M. Brittle fracture of non-ceramic insulators. In: The...
  • Kumosa et al. M. Analytical and experimental studies of substation NCIs. Final report to the Bonneville Power...
  • Cited by (62)

    • Impact of environmental conditions on the tribological performance of polymeric composites

      2022, Tribology of Polymers, Polymer Composites, and Polymer Nanocomposites
    • Effective through-the-thickness diffusivity of plain-woven composite from analytical homogenization

      2021, Composites Science and Technology
      Citation Excerpt :

      Consequences of moisture in the matrix material include plasticization and decrease in the glass transition temperature [10], swelling, i.e. a volumetric change in the polymer that can cause cracks and failure, hydrolysis where reactions between water and the polymer cause ruptures of the chemical bonds [11]. Glass fibres are vulnerable to stress corrosion in which exposure to tension and moisture leads to degradation and loose of strength [12]. Finally, the fiber-matrix interface can act as a hygroscopic material leading to higher moisture concentrations than the surrounding materials.

    • What happens to glass fiber under extreme chemical conditions?

      2020, Journal of Non-Crystalline Solids
      Citation Excerpt :

      In such cases, corrosion-induced strength degradation in FRPs causes the material to fail at a much lower stress than its ultimate strength [9,10]. It is believed that one of the obstacles preventing the more extensive use of FRP structures is the lack of long-term durability and performance during service in critical environments [11,12]. Consequently, it is necessary to understand the behavior of these materials under extreme chemical conditions [13].

    View all citing articles on Scopus
    View full text