Preservation of Evidence. Whether or not an on-site examination has been done, the samples should be handled so that maximum information can be gained before any sample is damaged, destroyed, or contaminated, preventing further tests Also, a complete written and photographic record should be kept through all investigation stages Visual Examination and Cleaning. First, the sample is examined visually, most often with the aid of a low-power hand magnifier. At this stage, important features include the extent of damage, general appearance of the damage zone, and the color, texture, and quantity of surface residues. If substantial amounts of foreign matter are visible, cleaning is necessary before further examination. The residues can be removed in some areas, leaving portions of the failure region in the as- received condition to preserve evidence. When only small amounts of foreign matter are present, cleaning can be deferred so that the surface can be examined with a stereomicroscope before and after cleaning. Cleaning also can be deferred until necessary for surface examination at higher magnifications or for the preparation of metallographic specimens. Small amounts of residues can be removed using transparent tape or acetate replicas and retained for later analysis ashing with water or solvent, with or without the aid of an ultrasonic bath, usually adequately removes soft residues that obscure the view. Inhibited pickling solutions will remove adherent rust or scale. Usually, the cleaning solutions should be saved for later analysis and identification of the substance removed Nondestructive Tests. For parts in which internal damage may have resulted from corrosion or from the combined effects of corrosion, stress, and imperfections, nondestructive testing is desirable. Radiography and ultrasonics can be used to locate internal discontinuities, and magnetic-particle and liquid-penetrant techniques are used to locate surface imperfections. Methods such as eddy current and holography are used less frequently, mainly because these methods require the use of standards to accurately interpret the data Microscopic Examination. Examination by both light microscopy and electron microscopy can be used to observe minute features on corroded surfaces, to evaluate microstructure of the metallic parts, and to observe the manner and extent to which the metal was attacked by the corrodent Viewing the cleaned surface with a stereomicroscope clearly shows gross topographic features such as pitting, cracking, or surface patterns that can provide information about the failure mechanism. This information includes whether corrosion was the sole phenomenon involved, the type of corrosion, and whether other mechanisms, such as wear and fracture, also were operative If the features cannot be observed clearly using a stereomicroscope, instruments such as deep-field photographic microscopes or a SEM may be used. These instruments produce images with a greater depth of field and, therefore, can resolve the topographical features of very rough surfaces. Transmission electron microscopy, using replicas, can resolve extremely fine features Microscopic examination of polished or polished-and-etched specimens can reveal microstructural features as well as damage such as cracking. If the corrosion products possess sufficient coherence and hardness to be polished, they should be retained. One way to keep the surface material in place is to impregnate the sample with a casting-type resin and allow it to harden before cutting samples. Usually, it is helpful to vacuum impregnate the sample during the casting process to be sure that any surface-connected voids are filled with casting plastic. To secure maximum quality of retention, polishing on napless cloths with diamond abrasives is used Chemical analysis of the Corrosion Products The job of the failure analyst is to establish what role, if any, the corrosion products played in the failure and to identif and analyze the metal or metals of which the failed part was made, the environment to which the failed part was exposed inhomogeneities in the part surface, and foreign matter and metal surfaces Both conventional techniques(such as wet chemical analysis, emission spectroscopy, x-ray diffraction, infrared spectrophotometry, gas chromatography, and x-ray fluorescence spectrography) and special techniques(such as energy dispersive x-ray spectrometry, electron-microprobe analysis, ion-microprobe analysis, Auger-electron spectrometry Mossbauer spectrometry, and electron diffraction) may be needed to define the composition and structure of various substances completely Identification and analysis of the metals of which the failed part was made usually is routine. Ordinarily, the purpose is not to look for minor deviations from the chemical composition specified for the part, but rather to check for possible major deviations in composition and to make sure that the correct alloy was used. However, for some austenitic stainless steels, the corrosion resistance and other properties of welded joints require close control over the composition of the stainless steel and the weld metal The bulk composition of the failure environment as well as the local composition of the environment at the metal interface are important in determining whether, or how, corrosion contributed to failure. Composition of the environment usually can be obtained chemically or spectroscopically Serious corrosion damage can result from the presence of inhomogeneities in the surface of a metal part used in a corrosive environment. A classic example is severe local attack on a stainless steel because of embedded particles of tramp" iron in the surface of the stainless steel. Establishing the nature of a layer of material on the metal surface may be more difficult than analyzing the material, especially when a thorough on-site investigation cannot be made. The surfacPreservation of Evidence. Whether or not an on-site examination has been done, the samples should be handled so that maximum information can be gained before any sample is damaged, destroyed, or contaminated, preventing further tests. Also, a complete written and photographic record should be kept through all investigation stages. Visual Examination and Cleaning. First, the sample is examined visually, most often with the aid of a low-power hand magnifier. At this stage, important features include the extent of damage, general appearance of the damage zone, and the color, texture, and quantity of surface residues. If substantial amounts of foreign matter are visible, cleaning is necessary before further examination. The residues can be removed in some areas, leaving portions of the failure region in the as￾received condition to preserve evidence. When only small amounts of foreign matter are present, cleaning can be deferred so that the surface can be examined with a stereomicroscope before and after cleaning. Cleaning also can be deferred until necessary for surface examination at higher magnifications or for the preparation of metallographic specimens. Small amounts of residues can be removed using transparent tape or acetate replicas and retained for later analysis. Washing with water or solvent, with or without the aid of an ultrasonic bath, usually adequately removes soft residues that obscure the view. Inhibited pickling solutions will remove adherent rust or scale. Usually, the cleaning solutions should be saved for later analysis and identification of the substance removed. Nondestructive Tests. For parts in which internal damage may have resulted from corrosion or from the combined effects of corrosion, stress, and imperfections, nondestructive testing is desirable. Radiography and ultrasonics can be used to locate internal discontinuities, and magnetic-particle and liquid-penetrant techniques are used to locate surface imperfections. Methods such as eddy current and holography are used less frequently, mainly because these methods require the use of standards to accurately interpret the data. Microscopic Examination. Examination by both light microscopy and electron microscopy can be used to observe minute features on corroded surfaces, to evaluate microstructure of the metallic parts, and to observe the manner and extent to which the metal was attacked by the corrodent. Viewing the cleaned surface with a stereomicroscope clearly shows gross topographic features such as pitting, cracking, or surface patterns that can provide information about the failure mechanism. This information includes whether corrosion was the sole phenomenon involved, the type of corrosion, and whether other mechanisms, such as wear and fracture, also were operative. If the features cannot be observed clearly using a stereomicroscope, instruments such as deep-field photographic microscopes or a SEM may be used. These instruments produce images with a greater depth of field and, therefore, can resolve the topographical features of very rough surfaces. Transmission electron microscopy, using replicas, can resolve extremely fine features. Microscopic examination of polished or polished-and-etched specimens can reveal microstructural features as well as damage such as cracking. If the corrosion products possess sufficient coherence and hardness to be polished, they should be retained. One way to keep the surface material in place is to impregnate the sample with a casting-type resin and allow it to harden before cutting samples. Usually, it is helpful to vacuum impregnate the sample during the casting process to be sure that any surface-connected voids are filled with casting plastic. To secure maximum quality of retention, polishing on napless cloths with diamond abrasives is used. Chemical Analysis of the Corrosion Products The job of the failure analyst is to establish what role, if any, the corrosion products played in the failure and to identify and analyze the metal or metals of which the failed part was made, the environment to which the failed part was exposed, inhomogeneities in the part surface, and foreign matter and metal surfaces. Both conventional techniques (such as wet chemical analysis, emission spectroscopy, x-ray diffraction, infrared spectrophotometry, gas chromatography, and x-ray fluorescence spectrography) and special techniques (such as energy dispersive x-ray spectrometry, electron-microprobe analysis, ion-microprobe analysis, Auger-electron spectrometry, Mössbauer spectrometry, and electron diffraction) may be needed to define the composition and structure of various substances completely. Identification and analysis of the metals of which the failed part was made usually is routine. Ordinarily, the purpose is not to look for minor deviations from the chemical composition specified for the part, but rather to check for possible major deviations in composition and to make sure that the correct alloy was used. However, for some austenitic stainless steels, the corrosion resistance and other properties of welded joints require close control over the composition of the stainless steel and the weld metal. The bulk composition of the failure environment as well as the local composition of the environment at the metal interface are important in determining whether, or how, corrosion contributed to failure. Composition of the environment usually can be obtained chemically or spectroscopically. Serious corrosion damage can result from the presence of inhomogeneities in the surface of a metal part used in a corrosive environment. A classic example is severe local attack on a stainless steel because of embedded particles of “tramp” iron in the surface of the stainless steel. Establishing the nature of a layer of material on the metal surface may be more difficult than analyzing the material, especially when a thorough on-site investigation cannot be made. The surface