Fractography of Metals and Plastics

7
16 Practical Failure Analysis Volume 2(5) October 2002 Fractography of Metals and Plastics by Ronald J. Parrington, P.E. T U T O R I A L Historical Perspective Plastics have been in existence for approximately 130 years. John Hyatt patented nitrocellulose, the first commercial plastic, in 1869. However, full-scale development and use of plastics is only approximately 50 years old. In contrast, metals have been in use for hundreds of years. The application of engineering mat- erials is unavoidably accompanied by the occurrence of failures, many of which have been catastrophic. The consequences of material failures, in- cluding deaths, financial losses, and legal ramifications, have encouraged the development of effective failure analysis methods. Although the cost of failure analysis may exceed the value of the part, the cost of service failures usually far exceeds the cost of failure analysis. Many of the tech- niques used over the years for the eval- uation of metals have been success- fully applied to plastics, with only minor modifications. Fractography is arguably the most valuable tool available to the failure analyst. Fractography, a term coined in 1944 to describe the science of ex- amining fracture surfaces, has actually been used for centuries as part of the field of metallurgy. Even before that, however, Stone Age man possessed a working knowledge of fracture. Arch- eological findings of lithic imple- ments, weapons, and tools shaped from stone by controlled fracture indicate that prehistoric man knew how to select rocks with favorable fracture behavior, use thermal spalling to detach bedrock from the working core, and shape stone by pressure flaking. Fractography, as we know it today, developed in the 16th century as a quality-control practice employed for ferrous and nonferrous metalworking. De La Pirotechnia, published by Van- noccio Biringuccio in 1540, [1] is one of the first documents to detail frac- tographic techniques. F ractography is critical to failure analysis of metals and plastics. Fractography of plastics is a rel- atively new field with many similarities to metals. Using case histories, various aspects of failure analysis and fractography of metals and plastics are compared and contrasted. Failure modes common to both metals and plastics include ductile overload, brittle fracture, impact, and fatigue. Analogies can also be drawn between stress-corrosion cracking (SCC) of metals and stress cracking of polymers. Other metal/plastic failure analogies include corro- sion/chemical aging, dealloying/ scission, residual stress/frozen-in stress, and welds/knit lines. Stress raisers, microstructure, material defects, and thermomechanical history play important roles in both types of materials. The key fractographic features for metals and plastics are described in this paper.

description

Fractography of Metals and Plastics by Parrington, Ronald J

Transcript of Fractography of Metals and Plastics

  • (continued)Fractography of Metals and Plastics

    16 Practical Failure AnalysisVolume 2(5) October 2002

    Fractography of Metals and Plastics

    by Ronald J. Parrington, P.E.T U T O R I A L

    Historical PerspectivePlastics have been in existence for

    approximately 130 years. John Hyattpatented nitrocellulose, the firstcommercial plastic, in 1869. However,full-scale development and use ofplastics is only approximately 50 yearsold. In contrast, metals have been inuse for hundreds of years.

    The application of engineering mat-erials is unavoidably accompanied bythe occurrence of failures, many ofwhich have been catastrophic. Theconsequences of material failures, in-cluding deaths, financial losses, andlegal ramifications, have encouragedthe development of effective failure

    analysis methods. Although the costof failure analysis may exceed thevalue of the part, the cost of servicefailures usually far exceeds the costof failure analysis. Many of the tech-niques used over the years for the eval-uation of metals have been success-fully applied to plastics, with onlyminor modifications.

    Fractography is arguably the mostvaluable tool available to the failureanalyst. Fractography, a term coinedin 1944 to describe the science of ex-amining fracture surfaces, has actuallybeen used for centuries as part of thefield of metallurgy. Even before that,however, Stone Age man possessed a

    working knowledge of fracture. Arch-eological findings of lithic imple-ments, weapons, and tools shaped fromstone by controlled fracture indicatethat prehistoric man knew how to selectrocks with favorable fracture behavior,use thermal spalling to detach bedrockfrom the working core, and shape stoneby pressure flaking.

    Fractography, as we know it today,developed in the 16th century as aquality-control practice employed forferrous and nonferrous metalworking.De La Pirotechnia, published by Van-noccio Biringuccio in 1540,[1] is oneof the first documents to detail frac-tographic techniques.

    F ractography is critical to failure analysis of metals and plastics. Fractography of plastics is a rel-atively new field withmany similarities tometals. Using case histories,various aspects of failure analysisand fractography of metals andplastics are compared andcontrasted.

    Failure modes common toboth metals and plastics includeductile overload, brittle fracture,impact, and fatigue. Analogiescan also be drawn betweenstress-corrosion cracking (SCC)of metals and stress cracking ofpolymers. Other metal/plasticfailure analogies include corro-sion/chemical aging, dealloying/scission, residual stress/frozen-instress, and welds/knit lines. Stress raisers, microstructure, material defects, and thermomechanical historyplay important roles in both types of materials. The key fractographic features for metals and plasticsare described in this paper.

  • 17Practical Failure Analysis Volume 2(5) October 2002

    Invention of the optical microscopein 1600 provided a significant newtool for fractography, yet it was notused extensively by metallurgists untilthe eighteenth century. In 1722, R.A.de Raumur[2] published a book withengravings that depicted macroscopicand microscopic fracture surfaces ofiron and steel. Interestingly, the cate-gories of macroscopic features devel-oped by de Raumur have remainedessentially unchanged through thecenturies.

    Partly due to the development ofmetallographic techniques for exam-ining cross sections of metals, interestin microfractography waned duringthe nineteenth century. Metalworkerscontinued to use fractographic tech-niques for quality-assurance purposes,but, for the most part, researchers andpublications ignored fractography.

    Several technological developmentsin the twentieth century revitalizedinterest in fractography. Carl A.Zapffe[3] developed and extensivelyused fractographic techniques tostudy the hydrogen embrittlement ofsteels. His work led to the discoveryof techniques for photographingfracture surfaces at high magnifi-cations. The first fractographs werepublished by Zapffe in 1943.

    An even more revolutionary devel-opment was the invention of thescanning electron microscope (SEM).The first SEM appeared in 1943.Unlike the transmission electron mi-croscope, which was developed a fewyears earlier, it could be used for frac-ture surface examination. An SEMwith a guaranteed resolution of ap-proximately 500 became commer-cially available in 1965. Comparedwith the optical microscope, the SEMexpands resolution by more than oneorder of magnitude and increases thedepth of focus by more than two

    orders of magnitude. The tools formodern fractography were essentiallyin place before plastics achieved wide-spread use.

    Failure Analysis OverviewThe general procedure for conduct-

    ing a sound failure analysis is similarfor metallic and nonmetallic mater-ials. The steps include: (1) informa-tion gathering; (2) preliminary, visualexamination; (3) nondestructive test-ing; (4) characterization of materialproperties through mechanical, chem-ical, and thermal testing; (5) selection,preservation, and cleaning of fracturesurfaces; (6) macroscopic examinationof fracture surfaces, secondary crack-ing, and surface condition; (7) micro-scopic examination; (8) selection,preparation, and examination of crosssections; (9) identification of failuremechanisms; (10) stress/fracturemechanics analysis; (11) testing tosimulate failure; and (12) data review,formulation of conclusions, and re-porting.

    Although the basic steps of failureanalysis are nearly identical, somedifferences exist between metals andplastics. Nondestructive testing ofmetals includes magnetic-particle,eddy-current, and radiographic in-spection methods that are not gener-ally applicable to plastics, forobvious reasons. However,ultrasonic and acoustic em-ission techniques find ap-plications for both materials.

    Similarly, different chemicaltest methods are necessary.Typical test methods formetals are optical emissionspectrometry, inductivelycoupled plasma, and combus-tion. Fourier transform in-frared spectroscopy is usedextensively to identify plastics

    by molecular bonding, and thermaltesting, differential scanning calori-metry, and thermogravimetric analysisare also very important for polymercharacterization. Energy-dispersive x-ray spectroscopy, used in conjunctionwith the SEM, is a very practical toolfor elemental chemical analysis ofboth metals and plastics. Also note-worthy is that different chemical solu-tions are required for metals and plas-tics to clean and/or protect fracturesurfaces and to etch cross sections toreveal microstructure.

    Causes of FailureOf course, the primary objective of

    a materials failure analysis is to de-termine the root cause of failure.Whether dealing with metallic ornonmetallic materials, normally, theroot cause can be assigned to one offour categories: design, manufactur-ing, service, or material. Often, severaladverse conditions contribute to thepart failure. Many of the potentialroot causes of failure are common tometallic and nonmetallic materials.

    Improper materials selection, overlyhigh stresses, and stress concentrationsare examples of design-relatedproblems that can lead to prematurefailure. Materials selection must takeinto account environmental sensitivi-

    Fig. 1 Fracture of a glass-filled polyamide threaded partdue to stress concentration at the thread root

  • (continued)Fractography of Metals and Plastics

    18 Practical Failure AnalysisVolume 2(5) October 2002

    ties as well as requisite mechanicalproperties and welding/joining char-acteristics. Stress raisers are frequentlya preferred site for fracture origin,

    Fig. 4 Microbiologically induced corrosion of a 304 SSTvessel weld, characterized by pitting and selectiveleaching (arrow)

    Fig. 5 Hollowing out of a polyacetal hinge due to acid-catalyzed hydrolysis

    Fig. 2 Cross section showing fracture along the knit line ofa perfluoralkoxyethylene-lined impeller

    Fig. 3 Cross section of a polyacetal hinge that fractured(arrow) through an area of porosity

    particularly in fatigue. Stress raisersinclude thread roots (Fig.1), sharpradii of curvature, through holes, andsurface discontinuities (e.g., gatemarks in molded plastic parts).

    Similarly, many manufac-turing and material problemsfound in metals also areobserved or have a corollaryin plastics. Weldments are atrouble-prone area for metals,as are weld lines or knit linesin molded plastics (Fig. 2).High residual stresses canresult from metalforming,heat treatment, welding, andmachining. Similarly, highfrozen-in stresses in in-jection-molded plastic partsoften contribute to failure.Porosity and voids are com-mon to metal castings andplastic molded parts (Fig. 3).Pores and voids serve asstress raisers and reduceload-carrying capability.Other manufacturing- andmaterial-related problemsthat may lead to failureinclude adverse thermomech-anical history, poor micro-structure, material defects,and contamination.

    Environmental degradation is oneof the most important service-relatedcauses of failure for metals and plas-tics. Other degradation processes in-clude excessive wear, impact, overload-ing, and electrical discharge.

    Failure MechanismsAnother key objective of failure anal-

    ysis is to identify the failure mechan-ism(s). Once again, some failure modesare identical for metals and plastics.These modes include ductile overload,brittle fracture, impact, fatigue, wear,and erosion.

    Analogies also can be drawn be-tween metals and plastics with regardto environmental degradation.Whereas metals corrode by an elec-trochemical process, plastics arevulnerable to chemical changes fromaging or weathering. Stress-corrosioncracking (SCC), a specific form ofmetallic corrosion, is similar in manyways to stress cracking of plastics.Both result in brittle fracture due tothe combined effects of tensile stressand a material-specific aggressive en-vironment. Similarly, dealloying orselective leaching in metals (Fig. 4),the preferential removal of oneelement from an alloy by corrosion,is somewhat similar to scission of

    polymers (Fig. 5), aform of aging thatcan cause chemicalchanges by selectivelycutting molecularbonds.

    Analogies can alsobe drawn betweenmetals and anothertype of polymer:rubber. Internal hy-drogen in steels canprecipitate and causehydrogen damage,which is frequently

  • 19Practical Failure Analysis Volume 2(5) October 2002

    (continued on page 44)

    characterized by localized brittle areasof high reflectivity, known as flakesor fisheyes, on otherwise ductile frac-

    ture surfaces (Fig. 6). Similarly, explo-sive decompression in rubber O-ringsproduces fisheye-like ovular patterns

    on the fracture surfaces (Fig.7). Explosive decompression isthe formation of small rupturesor embolisms when an elasto-meric seal, saturated with high-pressure gas, experiences anabrupt pressure reduction.This failure mechanism isanalogous to the bends thatafflict divers who surface tooquickly.

    FractographyWhen material failure in-

    volves actual breakage, frac-tography can be employed toidentify the fracture origin,direction of crack propa-gation, failure mechanism,material defects, environmen-tal interaction, and the natureof stresses. Some of the mac-roscopic and microscopicfeatures employed by thefailure analyst to evaluatefracture surfaces of metalsand plastics are describedsubsequently. Note, however,that many of the fracto-graphic features described forplastics are not observable for

    reinforced plastics and plasticscontaining high filler content.

    Macroscopically VisibleFractographic Features

    On a macroscopic scale, all fractures(metals and plastics) fall into one oftwo categories: ductile and brittle.Ductile fractures are characterized bymaterial tearing and exhibit grossplastic deformation. Brittle fracturesdisplay little or no macroscopicallyvisible plastic deformation andrequire less energy to form. Ductilefractures occur as the result of appliedstresses exceeding the material yieldor flow stress. Brittle fractures mayoccur at stress levels below the materialyield stress. In practice, ductile frac-tures occur due to overloading orunderdesigning and are rarely thesubject of a failure analysis. However,the unexpected brittle failure of nor-mally ductile materials is frequentlythe subject of a failure analysis.

    Many macroscopically visible frac-tographic features serve to identifythe fracture origin(s) and direction ofcrack propagation. Fractographic fea-tures common to metals and plasticsare radial marks and chevron patterns.Radial marks (Fig. 8) are lines on a frac-ture surface that radiate outward fromthe origin and are formed by the in-

    tersection of brittlefractures propagatingat different levels.Chevron or herring-bone patterns areactually radial marksresembling nestedletter Vs and point-ing toward the origin.

    Fatigue failures inmetals display beachmarks and ratchetmarks that serve to

    Fig. 6 Hydrogen damage of induction-hardened steelpiston rod displaying fisheyes

    Fig. 7 Explosive decompression fracture of rubber O-ring,characterized by fisheye-like patterns

    Fig. 8 Beach and radial marks visible on torsional fatiguefracture of a 6 in. diameter 4340 shaft

    Fig. 9 Brittle fracture of an epoxy layer displaying amirror zone, rib marks, and hackles

  • (continued)Fractography of Metals and Plastics

    44 Practical Failure AnalysisVolume 2(5) October 2002

    Fig. 13 Intergranular fracture of an embrittled cast steelpneumatic wrench

    identify the origin and the failuremode. Beach marks (Fig. 8) are mac-roscopically visible semielliptical linesrunning perpendicular to the overalldirection of fatigue crack propagationand marking successive positions ofthe advancing crack front. Ratchetmarks are macroscopically visible linesrunning parallel to the overalldirection of crack propagation andformed by the intersection of fatiguecracks propagating from multipleorigins.

    Brittle fractures in plastics also ex-hibit characteristic features, several ofwhich are macroscopically visible (Fig.9). These features may include a mir-ror zone at the origin, a mist region,and rib marks. The mirror zone is aflat, featureless regionsurrounding theorigin and associatedwith the slow crack-growth phase offracture. The mistregion is locatedimmediately adjacentto the mirror zoneand displays a mistyappearance. This areais a transition zonefrom slow to fastcrack growth. Ribmarks are semi-elliptical lines re-sembling beachmarks in metallicfatigue fractures.

    MicroscopicallyVisibleFractographicFeatures

    On a microscopicscale, ductile fracturein metals (Fig. 10)displays a dimpledsurface appearancecreated by microvoid

    Fig. 10 Dimpled appearance typical of ductile fracture ofmetallic materials

    Fig. 11 Fracture of a polyethylene tensile-test specimenexhibiting material stretching

    Fig. 12 Brittle fracture of an FC-0205 powder metalcontrol rod displaying cleavage facets

    coalescence. Ductile fracture in plas-tics (Fig. 11) is characterized by mat-erial stretching related to the fibrillarnature of the polymer response tostress. Although a part may fail ina brittle manner, ductile fracturemorphology is frequently observedaway from the origin. For example,the final fast fracture by ductileoverload produces the shear lip inmany metal failures, even when thecrack originated and was propagatedby SCC, fatigue, or hydrogen em-brittlement processes. The extent ofthis overload region is an indicationof the stress level. Generally, the largerthe overload region, the higher thestress level on the failed component.

    Brittle fracture of metallic materials

    may result from numerous failuremechanisms, but there are only a fewbasic microfractographic features thatclearly indicate the failure mechan-ism. These features are cleavage facets(Fig. 12), intergranular facets (Fig.13), and striations (Fig. 14). Cleavagefacets form in body-centered cubic(bcc) and hexagonal close-packedmetals when the crack path follows awell-defined transgranular crystallo-graphic plane (e.g., the {100} planesin bcc metals). Cleavage is character-istic of transgranular brittle fracture.Intergranular fracture, recognizableby its rock candy appearance, occurswhen the crack path follows grainboundaries. Intergranular fracture istypical of many forms of SCC, hy-

    (continued from page 19)

  • 45Practical Failure Analysis Volume 2(5) October 2002

    drogen embrittlement, and temper-embrittled steel. Fatigue failures ofmany metals exhibit striations at highmagnifications. (Normally, magni-

    fications of 500 to 2,500 are re-quired.) Striations are semiellipticallines on a fatigue fracture surface thatemanate outward from the origin and

    mark the crack-front positionwith each successive stresscycle. The spacing of fatiguestriations is usually veryuniform and can be used tocalculate the crack growthrate, if the cyclic stressfrequency is known. Stria-tions are discriminated fromstriation-like artifacts on thefracture surface in that truefatigue striations never crossor intersect one another.

    Plastics do not displaycleavage and intergranularfracture. However, similar tometals, striations are foundon fatigue fracture surfaces(Fig. 15, 16). Striations inplastics typically are observ-able at much lower magnifi-cations (50 to 200). How-ever, local softening andmelting due to hystereticheating can obliterate fatiguestriations in less rigid plastics.

    In addition to mirrorzones, mist regions, and ribmarks, which are normallyvisible without the aid of amicroscope, brittle fracture ofplastics may display hackles,Wallner lines, and conicmarks. Hackles (Fig. 9) aredivergent lines radiatingoutward from the fractureorigin. They are analogous toriver patterns observed on thecleavage facets of trans-granular brittle fractures ofmetals. Wallner lines arefaint, striation-like markingsformed by the interaction ofstress waves reflected from

    Fig. 14 Fatigue striations visible on type 302 stainless steelspring fracture

    Fig. 15 Fatigue striations emanating from fracture origin ofpolycarbonate latch handle

    Fig. 16 SEM micrograph of fatigue striations shown inFig.15

    physical boundaries with the ad-vancing crack front. Conic marks areparabolic-shaped lines pointing backtoward the origin. Hackles andWallner lines may or may not be visi-ble without the aid of a microscope.

    Closing RemarksFractographic techniques, develop-

    ed and applied to metal failures forcenturies, have been readily adaptedto the fracture analysis of plasticssince their emergence as a key engi-neering material over the last 50 years.However, more work remains to bedone to advance fractography ofplastics. One notable area for researchis fracture analysis of composites,reinforced plastics, and plastics con-taining high filler content. Fracturesof these materials too often aredismissed as inherently lackingmeaningful fractographic features.Finally, there is a definite need for anauthoritative publication on fracturein plastics.

    AcknowledgmentsThe author gratefully acknowledges

    the contributions of Dave Christieand Steve Ruoff of IMR Test Labs.

    References1. V. Biringuccio: De La Pirotechnia, Venice,

    1540; see translation by C.S. Smith andM.T. Gnudi: The Pirotechnia of VannoccioBiringuccio, American Institute of Mining,Metallurgical and Petroleum Engineers,New York, 1942.

    2. R.A. de Raumur: LArt de Convertir le FerForg en Acier, et LArt dAdocir le Fer Fondu,(in French), Michel Brunet, Paris, 1722;see translation by A.G. Sisco: RaumursMemoirs on Steel and Iron, University ofChicago Press, 1956.

    3. C.A. Zapffe and G.A. Moore: Trans.AIME, 1943, 154, pp. 335-59.

    Selected References W. Brostow and R.D. Corneliussen: Failure

    of Plastics, Hanser Publishers, Munich, 1986.

  • (continued)Fractography of Metals and Plastics

    46 Practical Failure AnalysisVolume 2(5) October 2002

    T.J. Davies and I. Brough: General Practicein Failure Analysis, Metals Handbook (9th

    ed.), vol. 11, Failure Analysis and Prevention,American Society for Metals, Metals Park,OH, 1986.

    M. Ezrin: Plastics Failure Guide: Cause andPrevention, Hanser Publishers, New York,1996.

    F.R. Larson and F.L Carr: How FailuresOccurTopography of Fracture Surfaces,Source Book in Failure Analysis, AmericanSociety for Metals, Metals Park, OH, 1974.

    Metals Handbook (8th ed.), vol. 9, Fracto-graphy and Atlas of Fractographs, AmericanSociety for Metals, Metals Park, OH, 1974.

    U. Portugall and K. Steinlein: Prac. Met-allography, 1999, 36(8), pp. 446-62.

    Ronald J. Parrington, IMR TestLabs, 131 Woodsedge Drive, Lans-ing, NY 14882. Contact e-mail:[email protected].

    Failure Analysis Resources from ASM International

    Volume 19: Fatigue and Fracture1996 1,057 pages ISBN: 0-87170-385-8 ASM Publication

    This volume contains the essential data necessary for you to makeinformed decisions on alloy design and material selection to controlfatigue and fracture. This Handbook is especially useful in evaluatingtest data and helping you understand the key variables that affectresults.

    Sections include: Fatigue Mechanisms, Crack Growth, Testing,Engineering Aspects of Fatigue Life, Fracture Mechanics ofEngineering Materials, Fatigue and Fracture Control, Castings,Weldments, Wrought Steels, Aluminum Alloys, Titanium Alloys andSuperalloys, Other Structural Alloys, Solders, Advanced Materials.Appendices contain coverage of fatigue strength parameters and stress-intensity factors.

    Order #06197G-RPADPFAJ

    Price: $198.00 ASM Member $159.00

    ASM Handbooks in print and CD-ROM Volume 11: Failure Analysis and Prevention1986 843 pages ISBN: 0-87170-017-4 ASM Publication

    Provides coverage of every aspect of this critical fieldfrom areview of the engineering aspects of failure analysis to the major typesof failures to in-depth analyses of failures of specific components andassemblies. Also included are failure analysis of composites, ceramics,polymers, and integrated circuits. Detailed case histories guide youthrough the investigative procedures used and the corrective measuresimplemented.

    Sections include: Engineering Aspects of Failure and Failure Analysis,Failure Mechanisms and Related Environmental Factors, Products ofPrincipal Metalworking Processes, Manufactured Components andAssemblies, Engineered and Electronic Materials.

    Order #06366G-RPADPFAJ

    Price: $198.00 ASM Member $159.00

    Volume 12: Fractography1987 517 pages ISBN: 0-87170-018-2 ASM Publication

    Provides engineers with enhanced capability for recognizing andinterpreting the various features of a fracture, enabling you to performimproved failure analyses and to better determine the relationship ofthe fracture mode to the microstructure. The Atlas of Fractographscontains more than 1,300 fractographs invaluable for your understandingof the causes and mechanisms of the fracture of engineering materials.

    Sections include: History of Fractography, Modes of Fracture, Prep-aration and Preservation of Fracture Specimens, Photography ofFractured Parts and Fracture Surfaces, Visual Examination and LightMicroscopy, Scanning Electron Microscopy, Transmission ElectronMicroscopy, Quantitative Fractography, Fractal Analysis of FractureSurfaces.

    Order #06365G-RPADPFAJ

    Price: $198.00 ASM Member $159.00