The document discusses the effects of sulfur and chlorine impurities in liquid penetrant testing materials on austenitic stainless steels. Stress corrosion testing was conducted on AISI 304 stainless steel specimens that were solution annealed, sensitized, or sensitized and etched. Specimens were coated with various liquid penetrant, developer, emulsifier, and cleaner materials and exposed to a moist environment. Specimens coated with a cleaner containing high chlorine levels or a developer with low chlorine consistently showed stress corrosion cracking if sensitized or sensitized and etched. Other liquid penetrant materials tested did not cause detectable stress corrosion cracking. The testing evaluated the potential for liquid penetrant materials to induce stress corrosion cracking in aust
Copyright:
Attribution Non-Commercial (BY-NC)
Available Formats
Download as PDF, TXT or read online from Scribd
Liquid Penetrant System Chemistry and Effluent Waste: Hapter
The document discusses the effects of sulfur and chlorine impurities in liquid penetrant testing materials on austenitic stainless steels. Stress corrosion testing was conducted on AISI 304 stainless steel specimens that were solution annealed, sensitized, or sensitized and etched. Specimens were coated with various liquid penetrant, developer, emulsifier, and cleaner materials and exposed to a moist environment. Specimens coated with a cleaner containing high chlorine levels or a developer with low chlorine consistently showed stress corrosion cracking if sensitized or sensitized and etched. Other liquid penetrant materials tested did not cause detectable stress corrosion cracking. The testing evaluated the potential for liquid penetrant materials to induce stress corrosion cracking in aust
The document discusses the effects of sulfur and chlorine impurities in liquid penetrant testing materials on austenitic stainless steels. Stress corrosion testing was conducted on AISI 304 stainless steel specimens that were solution annealed, sensitized, or sensitized and etched. Specimens were coated with various liquid penetrant, developer, emulsifier, and cleaner materials and exposed to a moist environment. Specimens coated with a cleaner containing high chlorine levels or a developer with low chlorine consistently showed stress corrosion cracking if sensitized or sensitized and etched. Other liquid penetrant materials tested did not cause detectable stress corrosion cracking. The testing evaluated the potential for liquid penetrant materials to induce stress corrosion cracking in aust
Copyright:
Attribution Non-Commercial (BY-NC)
Available Formats
Download as PDF, TXT or read online from Scribd
Download as pdf or txt
0 ratings0% found this document useful (0 votes)
75 views0 pages
Liquid Penetrant System Chemistry and Effluent Waste: Hapter
The document discusses the effects of sulfur and chlorine impurities in liquid penetrant testing materials on austenitic stainless steels. Stress corrosion testing was conducted on AISI 304 stainless steel specimens that were solution annealed, sensitized, or sensitized and etched. Specimens were coated with various liquid penetrant, developer, emulsifier, and cleaner materials and exposed to a moist environment. Specimens coated with a cleaner containing high chlorine levels or a developer with low chlorine consistently showed stress corrosion cracking if sensitized or sensitized and etched. Other liquid penetrant materials tested did not cause detectable stress corrosion cracking. The testing evaluated the potential for liquid penetrant materials to induce stress corrosion cracking in aust
Copyright:
Attribution Non-Commercial (BY-NC)
Available Formats
Download as PDF, TXT or read online from Scribd
Download as pdf or txt
You are on page 1of 0
Vilma G.
Holmgren, Magnaflux Division of Illinois Tool
Works, Glenview, Illinois Bruce C. Graham, Arlington Heights, Illinois Samuel J. Robinson, Sherwin Incorporated East, Burlington, Kentucky J. Thomas Schmidt, J.T. Schmidt Associates, Crystal Lake, Illinois Amos G. Sherwin, Sherwin Incorporated, South Gate, California Jack C. Spanner, Sr., Spanner Engineering, Richland, Washington 10 C H A P T E R Liquid Penetrant System Chemistry and Effluent Waste Influence of Sulfur and Halogens (Chlorides) in Liquid Penetrant Testing Materials Considerable concern in the nondestructive testing field is directed toward the effects of sulfur and halogens, principally chlorides, present in small amounts in liquid penetrant testing materials. This is largely attributable to various high temperature and exotic alloys such as nickel base alloys, austenitic stainless steels and titanium in the aerospace and nuclear industries. Even though liquid penetrants and processing materials are removed following testing, residues may be retained in crevices, joints and blind holes or other inaccessible areas. With inadequate cleaning, such residues may react detrimentally with the alloy surface after the components are placed in service. Sources of Sulfur and Halogen in Liquid Penetrant Testing Materials There are several possible sources for the presence of sulfur and halogen in liquid penetrant products. Raw materials from which liquid penetrant products are formulated may contain either or both of these ingredients as constituents or as trace contaminants. Where extremely small amounts of these ingredients are of concern, contamination from a high sulfur or high halogen atmosphere may contribute trace amounts of sulfur or halides. There is considerable difference of opinion as to the allowable limits of these contaminants. Paradoxically, there are numerous sources of sulfur and halide contamination that far overshadow the amounts present in liquid penetrant products, as in the following examples. 1. In certain techniques, precleaning takes place before liquid penetrant testing. Conceivably, residues from these operations could contribute significant residues. 2. Other in-process sources of contamination include cutting fluids or drawing and stamping oils. These fluids may contain high percentages of sulfur and/or chlorides. 3. Industrial waters may contain relatively high contents of chloride. Therefore cooling water, rinse water, makeup water for alkaline cleaners and other types of chemical processes may contribute either chlorides or sulfur residues. 4. In the case of jet engine components, ingestion of sea air or air containing high sulfur through industrial pollution can contribute very significantly to the amount of chlorides or sulfur gaining access to high temperature alloy parts. 5. Another possible source for sulfur contamination in jet engine components would be the jet engine fuel itself. All petroleum derivatives tend toward trace amounts of sulfur and, because of the high rates of throughput of jet engine fuel, the aggregate amount could be substantial. Specifications Restricting Halogen Content of Liquid Penetrant Testing Materials The continued incidence of stress corrosion cracking in austenitic stainless steels has focused increased attention on the halogen and sulfur content in liquid penetrant materials used for nondestructive testing. Restrictions are placed on the halogen and sulfur concentrations allowed in liquid penetrant materials intended for use on austenitic stainless steels. Influence of Sampling Procedures on Contamination Measurements It is important to recognize that, because of the nature of certain liquid penetrant materials, the sampling procedures can often influence the results of a given way of measuring chloride concentration. When fluorocarbon propellants and 288 Liquid Penetrant Testing PART 1. Effects of Sulfur and Chlorine Impurities in Liquid Penetrant Materials halogenated hydrocarbon solvents (such as 1,1,1-trichloroethane) were being used, they affected the results of chloride analysis, whether the test was performed on the whole sample or on the residue after evaporation of solvent or propellant. Most specifications still require residue analysis that gives more consistent and meaningful results. Significance of Postcleaning of Test Objects The purpose of the final postcleaning step in a liquid penetrant examination is to remove, as thoroughly as possible, all liquid penetrant materials applied during the test process. Thus, simple knowledge of the halogen content in liquid penetrant materials provides little useful information relative to the amount of halogen remaining on the part after liquid penetrant testing. On the other hand, even if all traces of halogen or sulfur bearing liquid penetrant materials are removed from the smooth surfaces of the part, the possibility of retaining damaging quantities of these compounds in macro (mechanical) or micro (intergranular) crevices should not be ignored. Stress Corrosion Testing of AISI 304 Austenitic Stainless Steel 1 The potential for liquid penetrant induced stress corrosion cracking in American Iron and Steel Institute (AISI) 304 stainless steel was investigated by coating U-bend test specimens with various liquid penetrant materials and exposing them to a moist 90 C (195 F) atmosphere for about three months. Test specimens, representing three different metallurgical conditions, were used to evaluate 19 different liquid penetrant, developer, emulsifier and cleaner materials having different halogen content and representing several manufacturers. Metallurgical Conditioning of AISI 304 Stainless Steel Specimens Test specimens for each of the following metallurgical conditions were used in the evaluation: (1) solution annealed, (2) sensitized and (3) sensitized and etched. The solution annealed specimens were included because AISI 304 stainless steel in this metallurgical condition is much less susceptible to halogen induced stress corrosion cracking than material in the sensitized condition. Specimens that were both sensitized and etched were included to examine the effect of microcrevices, which can conceivably retain appreciable quantities of liquid penetrant materials in spite of normal postexamination cleaning procedures. On completion of the metallurgical conditioning processes, the specimens were stressed using a stainless steel nut and bolt. Figure 1 illustrates the specimen configurations after forming and after stressing. Application of Liquid Penetrant Testing Materials to Test Specimens The final step in preparing the specimens for exposure testing was to apply the liquid penetrant materials. This was accomplished by immersing three stressed specimens from each of the three metallurgical groups in a given liquid penetrant material for about 6 h, after which the specimens were allowed to drain overnight in the open-end-down position. Water Vapor Exposure Conditions for Test Specimens Each specimen was hung, open end down, on a three layer rack formed from a single length of AISI 304 stainless steel tubing. The entire specimen and rack assembly was subjected to a warm, moist test atmosphere in an exposure chamber for a period of 80 days. A pool of distilled water was maintained in the bottom of the chamber and the temperature was maintained about constant at 90 C (195 F). Cold tap water was passed through the tubing rack to condense moisture on the specimen surfaces. 289 Liquid Penetrant System Chemistry and Effluent Waste FIGURE 1. Stainless steel U-bend test specimens after forming (left) and after stressing (right). Results of Stress Corrosion Exposure Tests of AISI 304 Stainless Steel The effect of the moist exposure conditions on test specimen integrity was evaluated by metallographic examination. First, several of the control specimens were sectioned and examined at 100 magnification as a check on the validity of the test parameters. Specimens in the sensitized and etched conditions, each of which had been coated with one of the liquid penetrant materials, were similarly examined for evidence of stress corrosion cracking. Examples of the metallographic examination results are shown in Fig. 2 to 6. Figures 2a and 2b are typical of the control specimens in the sensitized and in the sensitized and etched conditions, respectively. These uncontaminated specimens were completely unaffected during the 80 day exposure test, as were most of the coated specimens. Effects of Chlorine in Specific Liquid Penetrant Cleaner and Developer Intergranular cracking to about midthickness is evident in the AISI 304 stainless steel specimen which is shown in Fig. 3 and which was sensitized and coated with developer containing 0.2 percent chlorides. Figure 4 is typical of the three specimens that were sensitized and etched before being coated with the same developer. Figures 5 and 6 illustrate the severe intergranular and transgranular cracking that occurred when sensitized test specimens were coated with a cleaner containing about 80 percent chlorides. The specimen shown in Fig. 5 was in the sensitized condition whereas the specimen in Fig. 6 had been sensitized and etched to a depth of about 130 m (0.005 in.) before application of the cleaner. Overall Results of Stress Corrosion Tests on AISI 304 Stainless Steel The overall results of the moisture exposure tests on AISI 304 austenitic stainless steel specimens are summarized below. 1. Specimens coated with a cleaner that had a very high chloride content or a developer with low chloride content (see Figs. 3 to 6), consistently exhibited stress corrosion cracking after the 80 day exposure test when in the sensitized or sensitized and etched conditions. 2. None of the other liquid penetrant materials evaluated in this test caused detectable stress corrosion cracking. 3. Etching of the sensitized specimens did not appear to have reproducible effects on the severity of attack for the liquid penetrant materials that induced stress corrosion cracking. 290 Liquid Penetrant Testing (a) (b) FIGURE 2. Photomicrograph of sensitized AISI 304 stainless steel specimen (typical of five control specimens not coated with liquid penetrant test materials), after 80 day stress corrosion test in warm, moist atmosphere and after oxalic acid etch: (a) sensitized specimen; (b) etched specimen. 291 Liquid Penetrant System Chemistry and Effluent Waste FIGURE 3. Photomicrograph of sensitized specimen after 80 day exposure test and oxalic acid etch is typical of three specimens coated with developer. FIGURE 4. Photomicrograph of sensitized and etched specimen after 80 day exposure test and oxalic acid etch is typical of three specimens coated with developer. FIGURE 5. Photomicrograph of sensitized specimen after 80 day exposure test and oxalic acid etch is typical of three specimens coated with cleaner. FIGURE 6. Photomicrograph of specimen sensitized and etched to depth of about 130 m (0.005 in.) after 80 day exposure test and oxalic acid etch is typical of three specimens coated with cleaner. 4. Specimens in the solution annealed condition exhibited no evidence of stress corrosion attack. 5. Observations among triplicate specimens were fully reproducible. The results obtained during this study support the following conclusions. 1. The reproducibility of the exposure test results provides convincing evidence that certain liquid penetrant materials can induce stress corrosion cracking under the moderate exposure conditions investigated. 2. The determination methods that were developed offer an economic and reliable means of measuring chloride concentration in whole samples of liquid penetrant materials over the range of 10 to 1000 gg 1 . 3. Commercial liquid penetrant materials with total chloride concentrations below 1000 gg 1 are available. 4. Liquid penetrant materials from dichlorodifluoromethane (refrigerant-12) pressurized spray cans yield variable results when whole samples are analyzed, thus making it impossible to perform a reliable determination of chloride concentration. Specifying Liquid Penetrant Materials to Avoid Induced Stress Corrosion Cracking 1 It should be recognized that the cause of the stress corrosion cracking observed in the preceding tests is not actually known. These tests were designed only to examine the potential for liquid penetrant induced stress corrosion cracking under a specific set of exposure conditions and were not intended to isolate the cause of an attack that might be observed. Liquid penetrant materials that induced the observed stress corrosion cracking would qualify as acceptable materials in accordance with the halogen restrictions specified in applicable United States standards in effect from 1971 through 1981. It thus appears that restrictions on contaminant concentrations might be more appropriately specified for whole samples (including volatiles) instead of only for the residue, as is the prevalent practice. It also seems significant that, although most specifications restrict total halogens, none of the methods normally used to verify compliance (including those discussed herein) respond to fluoride. Thus, in actual practice, the most reactive of the halogen elements was tacitly ignored until 1981. Sources of Halogen Impurities in Liquid Penetrant Processing Materials Except for the certain cleaners and carriers that use halogenated hydrocarbon compounds, liquid penetrant materials do not usually contain halogens as constituent elements. Rather, the halogens occur as impurities in the bulk chemicals used to formulate these liquid penetrant materials. Preceding tests also indicate that commercial liquid penetrant materials can exhibit a rather wide variation in chloride content. It is also apparent that liquid penetrant materials quite low in hydrolyzable and even in total chloride content are available. The same statement can probably be made about fluoride content. Specification Restrictions on Sulfur Content of Liquid Penetrant and Leak Testing Materials For years, most new standards have included an upper limit for total sulfur impurities in expendable materials such as liquid penetrants, cleaners and developers; leak test fluids; coatings; adhesives; lubricants; and ultrasonic couplants. In contrast to the effect of halogens, sulfur and its compounds were not found to be significantly corrosive nor otherwise damaging to austenitic stainless steels at ambient temperatures. However, stress cracking does take place on high yield strength ferritic and austenitic steels when sulfides exist in the aqueous phase and on nickel and its alloys when sulfides are present during elevated temperature processing. Like halogens, sulfides have been related to the tendency toward hydrogen embrittlement. Investigators agree that minimizing residual amounts of sulfur in materials applied to iron base and nickel base alloys will reduce the likelihood of stress corrosion or sulfide stress cracking. Selection should be based on which product has the lowest sulfur content, as well as the lowest halogen content. The following information describes the action of sulfur compounds on different classes of structural materials. 292 Liquid Penetrant Testing Hydrogen Damage from Sulfur Compounds in Nonstainless Steels When an acid solution attacks a carbon or alloy steel surface, hydrogen is generated and deposited on the metal surface as an adsorbed film of atomic hydrogen. Normally, the film builds up and bubbles off the surface as molecular hydrogen gas. However, in the presence of even a few parts per million of hydrogen sulfide in the solution environment, much of the atomic hydrogen enters high strength steel. The presence of a sulfide in the steel appears to catalyze the generation of hydrogen. The formation of molecular hydrogen internally in voids produces high pressure, creates blisters on the surface and reduces the ductility. Delayed failures then can occur at stress levels well below the yield strength. The minimum yield strength level for sulfide stress cracking is about 600 MPa (90 000 lb f in. 2 ). The tendency for cracking is reduced as the pH rises from 3 to 6 (solutions become less acid). Inorganic and aliphatic organic sulfur compounds are generally more reactive than aromatic organics. Sulfide corrosion of nonstainless steel is reduced by the presence of halides in the low temperature aqueous phase. Corrosive Effects of Sulfur on Various Alloys High Yield Strength Stainless Steels The high yield strength AISI 635 stainless steel (17 percent chromium, 7 percent nickel) is very susceptible to cracking in a high sulfide environment (1000 gg 1 of hydrogen sulfide). Stainless steels with chromium contents less than 9 percent may corrode faster in a given sulfide concentration than carbon steel. AISI 300 Austenitic Stainless Steels Adequate protection against sulfide corrosion is provided by the lower strength standard AISI 300 series austenitic stainless steels with a basic alloy content of 18 percent chromium and 8 percent nickel. Iron Base Alloy with 14 Percent Chromium and 14 Percent Nickel The effect of sulfur as an impurity in a completely stable austenitic structure a high purity 14 percent chromium, 14 percent nickel, iron base material was evaluated in five such alloys with sulfur impurities: 80, 120, 240, 890 and 1700 gg 1 . Nonstress corrosion rates appeared to be directly related to sulfur content. Corrosion was uniform rather than localized. The extent of attack was slight compared with that found for commercial purity 14 percent chromium, 14 percent nickel, 72 percent iron alloys and of commercial, high temperature steels such as AISI 304. Stressed U-bend samples of 14 percent chromium, 14 percent nickel, iron base alloy representing the five levels of sulfur were exposed to 42 percent boiling magnesium chloride for 600 h. No evidence of cracking was observed. Similar alloys of commercial purity cracked after only 28 h exposure of the stressed samples to the strong chloride corrodent. It was concluded that, to produce excellent corrosion resistance to aggressive media, all impurities, rather than just sulfur, must be excluded from austenitic stainless steels. Although increasing additions of sulfur increased the corrosion rate slightly in the high purity austenitic matrix, the corrosion rate of the 1700 gg 1 or 0.17 percent sulfur alloy was much below that of commercial purity materials. Residual sulfur accumulations from liquid penetrant materials would not pose a special corrosion or stress corrosion problem at temperatures below 100 C (212 F) for parts produced from austenitic stainless steels. Nickel and Nickel Base Alloys Materials in which nickel predominates are subject to intergranular attack when they are heated in the presence of sulfur and sulfur compounds. Grain boundary stress corrosion cracking and nickel has occurred in sulfide environments above 643 C (1190 F), the melting point of the nickel/nickel sulfide eutectic. Fuels used in the heat treatment of nickel base alloy materials must have a low sulfur content. A marginal concentration of sulfur is stated to be 0.73 gm 3 of gases manufactured from coal and oil. Lubricants for deep drawing and spinning nickel alloys generally contain sulfur or lead and therefore must be removed before annealing. 293 Liquid Penetrant System Chemistry and Effluent Waste Environmental Conditions Affecting Intergranular Attack Intergranular attack by sulfur has been reported only for nickel base alloys subjected to elevated temperatures. However, it seems probable that austenitic stainless steels could also be adversely affected by exposure to sulfur or its compounds in its sensitizing temperature range of 590 to 650 C (1100 to 1200 F). At temperatures below 150 C (300 F), steels with a maximum yield strength of 520 MPa (75 000 lb f in. 2 ) should not be subject to sulfide stress cracking. The tendency for hydrogen absorption and embrittlement is reduced by raising the pH of the environment to 6 or greater. High yield strength austenitic such as 17 percent chromium, 7 percent nickel alloys should not be susceptible to cracking if the sulfide content of the environment is kept well below 1000 gg 1 . 294 Liquid Penetrant Testing Restrictions on Halogens in Expendable Test Materials Used on Austenitic Stainless Steels For years, it was common practice in the aerospace industry to use halogenated cleaners, acids and degreasers in the treatment of AISI 300 series stainless steels. However, pressure vessel codes and other standards in the United States prohibit expendable cleaners, liquid penetrants and processing materials containing more than one percent halogens. Investigations of cause and prevention of stress corrosion cracking of stainless steel have demonstrated that less than 50 gg 1 of certain hydrolyzable chlorides in combination with 1 to 200 gg 1 of oxygen can cause cracking under low stress and low temperature conditions. The likelihood of stress corrosion will diminish with decreases in concentration of halides and/or oxygen, with decreases in stress and with decreases in temperature. Laboratory specimens of stressed austenitic stainless steels failed when immersed in a 5 gg 1 chloride solution at 93 C (200 F) and in a 10 gg 1 chloride solution at 74 C (165 F). Because chloride concentrations in normal environments and residual stresses from fabrication and heat treatment generally exceed these levels, one might wonder at the usual long life and high performance characteristics of most stainless parts. There may be a relationship between hydrogen embrittlement and stress corrosion cracking. A further restriction of hydrogen generation in the right form and location could limit the chance for stress corrosion cracking to occur. The fact that many of the halogen compounds are volatile does not negate the potential danger. Cyclic evaporation and subsequent condensation is actually a means of concentrating the halogen in a restricted area such as a crack or crevice. Entrapment and contamination with surface adherent material can reduce the tendency of halogen compounds to become volatile. Mechanisms of Stress Corrosion Cracking of Austenitic Stainless Steels Austenitic stainless steels can fail by stress corrosion cracking, a brittle failure in ductile material which occurs as a result of the combined action of tensile stress and corrosion. Tensile stress may be externally applied or residual but it must be sustained or static. Where allowable, shot peening to alleviate surface tensile stresses will reduce the tendency for cracking. Although the minimum stress to initiate stress corrosion cracking will vary with the environment and with steel composition differences, stresses reportedly as low as 14 MPa (2000 lb f in. 2 ) have resulted in stress corrosion cracking of AISI 347 austenitic stainless steel. It would seem that significant stress concentrators would have to be operating to bring about such low stress failure in AISI 347 surrounded by vapors at 200 C (400 F) containing 50 gg 1 of chloride. Other Mechanisms of Cracking of Austenitic Stainless Steels Not all cracking occurring in fabrication or service is halide induced. Three other modes of failure exist that can take place at stress levels below the yield strength of austenitic stainless steel in a noncorrosive environment. 1. Fatigue failure may be induced thermally, mechanically or by vibration during operation. 2. Long time, elevated temperature stress rupture may occur. 3. Differences in coefficient of expansion and in thermal conductivity between cladding and base metal may lead to failure. In boiling water reactors, nonstress corrosion cracks were found to be longitudinal in cold worked tube and circumferential in solution annealed tubing. Although all potential causes should be considered in a failure investigation, good design should minimize the likelihood of cracking from mechanisms other than stress corrosion. 295 Liquid Penetrant System Chemistry and Effluent Waste PART 2. Mechanisms of Stress Corrosion Cracking of Austenitic Stainless Steels Fracture Paths Resulting from Stress Corrosion Both intergranular fracture and transgranular fracture may occur as a result of stress corrosion. Usually, cracking produced in stainless steel exposed to chlorides is transgranular. However, where there is a continuous grain boundary segregation, intergranular cracking will take place. There is a disposition toward a chemical concentration differential between grains and grain boundaries because impurities and alloy constituents tend to diffuse to the boundaries. Segregated solute such as silicon or phosphorus can occur in the grain boundaries of properly solution annealed AISI 304. On the other hand, a continuous grain boundary part of second phase, alloy carbides is formed by a sensitizing treatment at 663 C (1225 F) for 2 h followed by slow cooking. The likelihood of forming continuous carbide at the grain boundaries is a function of grain size and carbon percent. The tendency for grain boundary attack by a corrodent is independent of pH between 4.7 and 9.4. Surface pickling with boiling nitrate chromate can cause cracking only on sensitized material. Intergranular attack on an oxide free surface will only happen on stressed material or in crevices. Causes of Crack Initiation Plastic deformation may be a necessary condition for establishing a propensity for cracking. In austenitic stainless, even elastic strain is associated with a small amount of creep (plastic strain). Cracks can originate from elongated pits or from eroded grain boundaries. Reducing the temperature of the environment extends the life of the part before crack initiation but does not alter the minimum stress level requirements for cracking. Stress corrosion cracking always produces corrosion products. Apparently, reaction of the chloride on bare metal exposed at breaks in the protective oxide film on the stainless steel produces local anodic areas. The latter set up galvanic action leading to pitting and thence to crack initiation when under tensile stress. The minimum stress to initiate stress corrosion cracking depends on (1) the nature of the corrodent medium, (2) temperature, (3) anion availability, (4) history of the material and (5) availability of minimum amounts of chloride and oxygen or minimum amounts of hydroxide. The rate of corrosion is unaffected by varying the initial hydrogen ion concentration (pH) of the solution. The concentration effect of intermittent wetting and drying will shorten the time required for crack initiation. Under a constant stress level, cracking will occur most easily at 85 C (185 F), the maximum temperature for corrosion acceleration at a high enough level for crack initiation. The temperature limitation of 85 C provides a minimum driving force for progressive corrosion such that cracking must occur within 48 h or the rate of passive film repair will exceed the rate of penetration. Test for Susceptibility to Stress Corrosion The establishment of a standard corrosion medium could serve as a tool for measuring the susceptibility of material or parts to stress corrosion cracking. Because 875 gg 1 of sodium chloride vapors at 200 C (400 F) will produce severe cracking in a short time, this might be used as such a standard corrosion medium. Electrochemical Theory of Crack Propagation The electrochemical theory of stress corrosion assumes crack advancement by anodic dissolution of stained metal at the crack tip. It has been postulated that nitrogen may be responsible for the corrosion attack on strained, crack sensitive areas. Based on work with low nitrogen, 18 percent chromium, 8 percent nickel and 20 percent nickel alloys, investigators suggest that nitrogen, under conditions of lattice strain, readily diffuses to lattice imperfection sites. Nitride precipitation then forms a cathodic area that stimulates corrosion of the adjoining alloy, thereby favoring crack propagation. In nonsusceptible high nickel alloys, nickel apparently inhibits the formation of the cathodic precipitates. Electron microscopy has revealed what appear to be precipitated nitrides on slip planes. Slip increases the effective area of the cathodic nitride precipitates and thereby increases galvanic corrosion and aids crack propagation. Cracking may occur at anodic paths because of reaction of stainless steel with corrosion produced hydrogen. This hydrogen diffuses under an applied load to form stress oriented transgranular bands of hydrogen strained ferrite. Such transformed bands, highly anodic to the austenitic matrix, dissolve in a corrosive halide such as magnesium chloride to cause cracking. In nonsusceptible alloys, hydrogen precipitates as a hydride. 296 Liquid Penetrant Testing Some investigators suggest that cracking originates from a condition of reduced surface energy brought about by the adsorption of some ion species from the corrosive medium. Electrochemical processes might contribute to failure by removing existing films at localized areas that would otherwise prevent adsorption of the critical ion species or, at higher stress levels, might prevent healing of oxide films that become ruptured during loading. Electrochemical/Mechanical Theory of Crack Propagation The electrochemical/mechanical theory of stress corrosion presupposes a two stage repeating cycle consisting of a short period of chemical attack, which advances the crack very little but which triggers a sudden mechanical fracture over a longer distance and which is stopped apparently by a soft slip plane. Once a crack is initiated, it is not permanently arrested by contact with grain boundaries, twin planes, or inclusions, nor is its propagation influenced by changes in applied stress or by changes in the nature of the medium. At a low stress level, cracking occurs only in the area of the highest stress. Studies indicate that cracks extend in short bursts at average rates up to 4 mmh 1 in the condensing vapors above an 875 gg 1 sodium chloride solution at 204 C (400 F) or up to 10 mmh 1 , in a boiling magnesium chloride environment. Some investigators have recorded acoustic emission, suggesting association with crack extension. Other investigators were unable to detect acoustic emission during crack propagation. These observations suggest that very little acoustic emission activity accompanies intergranular stress corrosion cracking whereas the opposite is indicated for transgranular stress corrosion cracking. Perhaps high stacking fault energy, indicating tangled dislocations and easy cross slips in an alloy, inhibits cracking. Rapid stress corrosion crack propagation occurs in alloys having highly oriented dislocations (low stacking fault energies) and restricted slip. Effect of Wedging Action of Corrosion Products Deposition of corrosion products within progressing cracks may provide a wedging action that causes brittle fracture beyond the tip of the crack. Pressures in excess of 50 MPa (7000 lb f in. 2 ) have been measured for the wedging action of the corrosion product. It has been proposed that the chemical aspects of cracking are due to preferential attack on small quantities of transformed austenite at the tip of the crack or as a network of strain bands within the austenite grains. The trace of a crack on the surface does not necessarily indicate that the same propagation process fracture path has occurred in the bulk. The three dimensional path of a crack is determined mainly by the stress distribution near the advancing edge. This is not necessarily related to microstructural features (grain size and orientation) or crystallographic planes. Effect of Hydrogen Embrittlement on Stress Corrosion Cracking Hydrogen embrittlement must be assumed to have relation to stress corrosion cracking. It is known that protons (hydrogen nuclei) can move along with dislocations during plastic deformation because of the mobility of the proton. Hydrogen embrittlement can only occur when the proton diffusion rate is equal to or greater than the rate of movement of dislocations. Hydrogen collects preferentially in the tetrahedral lattice positions in the (112) planes, in the discontinuities associated with these planes and probably in the dislocations related to this crystallographic system. Movement of the most unstable dislocations can be accomplished in the presence of hydrogen by stresses well below the elastic limit, resulting in delayed failures. Stress Corrosion Immunity, Susceptibility and Prevention Six general rules postulated for susceptibility to stress corrosion are listed below. 1. A pure metal is immune to stress corrosion cracking. 2. Alloys made from pure metals may be susceptible. 3. Corroding conditions (media or environment) that produce cracking are specific to an alloy or alloy system. 4. Cathodic protection can prevent stress corrosion cracking or even stop crack propagation if applied while cracking is in process. 5. One or more minor impurity elements in a metal or alloy can affect its degree of immunity or susceptibility. 6. Changes in structure or homogeneity of an alloy by heat treatment can influence its immunity or degree of susceptibility. 297 Liquid Penetrant System Chemistry and Effluent Waste Effect of Cathodic Protection on Stress Corrosion Cracking A small amount of cathodic protection may inhibit stress corrosion cracking whereas a large amount may induce it. Arrangement of anodes in reactors, piping and heat exchangers for sufficient protection is frequently impossible and sacrificial anodes may be undesirable because of the corrosion products introduced into the coolant. Where the highly stressed component is the cathode, stress corrosion cracking may be stopped or restarted at will by applying an external potential. Conditions Conductive to Stress Corrosion Cracking Wherever combinations of high residual stress, crevices for entrapment of halides or hydroxides, heat transfer, evaporation and concentration occur, stress corrosion cracking is likely. Chlorides can also concentrate in aluminum corrosion products by an ion exchange mechanism. Thus, fabricated stainless steel parts that have been tested with liquid penetrants or cleaned by chlorinated solvents are prime targets for stress corrosion. Conditions That Inhibit Stress Corrosion Cracking Any approach to prevention or inhibition of stress corrosion cracking requires an understanding of the mechanism of cracking. A reduction of stress level will reduce the likelihood of rupture and will increase the life of the part. As previously stated, reducing the temperature will extend part life but will not alter the minimum stress level for cracking. Virtual elimination of either halides or oxygen from the environment (including liquid penetrant testing fluids) will prevent stress corrosion as long as the hydroxide concentration is under five percent. Cathodic protection will be an effective deterrent where it is practicable. Changes in design and operating conditions will be helpful if, for example, opportunities for fluid entrapment are reduced and operating temperatures and stress levels can be lowered. A substitution of less susceptible material such as carbon steel, nickel base alloys or corrosion resistant nickel-copper alloys for stainless steel is another preventive measure. Tensile stress relief by peening vibration or heat treatment can reduce the effect of a corrosive medium. Corrosive Environments The rate of corrosive attack on stressed stainless steel by various chlorides is essentially inversely proportional to their rate of attack on unstressed stainless. Information on the comparative effect of specific chlorides on stressed versus unstressed AISI 300 series stainless steels is given in Table 1. It would appear that stress corrosion requires selective localized attack, which can only be provided by corrodents that do not affect the pure alloy. Chloride Removal Following Liquid Penetrant Testing of Stainless Steels Stainless steel components should be as free as possible of surface residual chlorides after cleaning or liquid penetrant testing, particularly before heat treatment, welding or long term storage. A quantitative chloride determination will be required to establish the accuracy of a given flushing procedure. Acid treatment should be avoided to ensure that no intergranular corrosion produces a site for liquid sodium attack in a breeder reactor or for stress concentration. Final flushing after liquid penetrant testing of AISI 304 stainless steel can be done with alcohol, acetone or an aromatic hydrocarbon solvent such as benzene or toluene. The latter should provide effective removal of adsorbed and entrapped chlorides. However, all are highly flammable. Fluids for Cleaning of Stainless Steels Several chlorinated solvents such as nonflammable, low toxicity trichlorotrifluoroethane (refrigerant-113) can be used to clean austenitic stainless steels regardless of heat sensitization and degree of stress. The accompanying low 298 Liquid Penetrant Testing TABLE 1. Effects of chlorides on stainless steels. Chlorides General Attack Stress Corrosion Ammonium chloride pitting slow (long exposure) Calcium chloride nil rapid Chromium (III) chloride heavy nil Iron (III) chloride heavy nil Magnesium chloride nil rapid Mercury (II) chloride heavy nil Potassium chloride pitting slow (long exposure) Sodium chloride pitting slow (long exposure) Zinc chloride nil rapid boiling point of 45.8 C (114 F) facilitates complete removal by evaporation. 2 However, these solvents are no longer available for cleaning. They were regulated because of their high potential for ozone depletion. It has been reported that sulfamic acid, HSO 3 NH 2 , can be used for cleaning austenitic steel without causing stress corrosion cracking. A five percent solution of this acid at 70 C (160 F) can be applied for periods up to 24 h. No data have been reported to suggest that nonhydrolyzable halides such as fluorocarbons were associated with stress corrosion cracking. Superior resistance to stress corrosion has been reported for a high alloy, columbium stabilized, iron base, nickel-chromium-molybdenum austenitic stainless steel. This has a nominal analysis of 30 percent chromium, 34 percent nickel, 2.5 percent molybdenum and 3.5 percent copper. The high nickel content is credited with supplying the improved resistance to stress corrosion cracking Laboratory tests in the most severe corrodent, boiling magnesium chloride, show more than a tenfold increase in life for the columbium stabilized nickel-chromium-molybdenum austenitic stainless steel over AISI 304; however, the cost of the high alloy material is about four times that of AISI 304 for tubular products. Conclusions on Control of Stress Corrosion Cracking As long as the ASME Boiler and Pressure Vessel Code 3 limits total halogens in liquid materials for nondestructive testing methods to one percent, materials for use on austenitic stainless steel components should be selected for minimum halide content consistent with effectiveness and cost. The lower the halide residual concentration in the environment, the lower the possibility of stress corrosion. Thorough cleaning for removal of halide containing materials after their use will effectively eliminate any potential for stress corrosion. Stainless steels containing a nominal 8 percent nickel such as AISI 304, AISI 316 and AISI 347 have maximum sensitivity to chloride stress corrosion cracking. Attainment of minimum stress levels in stainless steel components is therefore desirable. Thermal, vibrational or peening stress relief can reduce residual stresses. Residual and/or applied tensile stresses as low as 14 MPa (2000 lb f in. 2 ) in combination with 74 C (165 F) or higher fluid environments containing 1 to 50 gg 1 or less of certain chlorides and 1 to 200 gg 1 or less of oxygen can result in stress corrosion cracking. Low concentrations of total halides (5 to 50 gg 1 ) cannot be accurately determined by current standard analytical procedures. Ion chromatography can analyze halide concentrations in this range (see ASTM E 165 4 for procedures). Final cleaning is recommended before heat treatment, forming, fabrication, storage or shipment to remove water, dissolve grease and oil, float away insoluble particulate without promoting stress corrosion cracking. 299 Liquid Penetrant System Chemistry and Effluent Waste Effects of Sulfur and Halogens on Liquid Penetrant Testing Materials Liquid penetrant testing is used on most nonferrous metals, particularly aluminum, magnesium, titanium, stainless steel and high nickel alloys. Stainless steels and titanium have been found to be subject to corrosion and embrittlement because of contact with chloride ions. High nickel alloys have been found subject to corrosion and embrittlement by contact with sulfur in the form of sulfide ions at elevated temperatures. Specifications Limiting Sulfur and Halogen in Liquid Penetrant Testing Materials Because of these findings, limits have been placed on the amount of sulfur and halogen that may be present in liquid penetrant testing materials used on high nickel alloys, stainless steel and titanium, particularly in components of nuclear power plants. The specifications have usually been for total sulfur content and total halogen content. Here it was assumed that the worst possible case would be where the liquid penetrant testing materials are not cleaned off the part after inspection and the test part is heated to temperature that will decompose the test materials, releasing the elements present for reaction with the metal of the part. Impurity Level Requirements for Liquid Penetrant Testing Materials Impurity level requirements for liquid penetrant testing materials are restricted by the ASME Boiler Code, Section V, Article 6. 3 This code limits maximum residual to 1 percent total sulfur and 1 percent total halogens as determined by ASTM D 129 5 and ASTM D 808 6 analytical techniques either/or ion chromotography, 4 when liquid penetrant testing materials are applied to nickel base alloys, austenitic stainless steels and titanium. To accommodate requirements that are more restrictive than those specified in the ASME Boiler Code, 3 consideration has been given to lower limits on impurities in expendable materials that will contact such metal surfaces. Contaminants levels are the result of impurities in the bulk chemical ingredients used to formulate liquid penetrant materials. Therefore, some variations may be observed between materials from different suppliers, between different types of materials from a single supplier and between samples of the same type of material purchased from a single supplier at different times. Specifications for Analysis for Sulfur or Halogens Analyses for sulfur or halogen impurities usually are conducted by the following methods although other methods are known to be equivalent or superior in accuracy, reproducibility and repeatability. 1. Analysis for water leachable chlorides is conducted according to ASTM D 2441 7 (modified by water reflux and potentiometric titration). 2. Analysis for halogenated compounds is conducted according to ASTM D 808. 6 Chloride compounds are absorbed in sodium carbonate (Na 2 CO 3 ) solution. The chloride content is determined by potentiometric titration. 3. Analysis for sulfur is conducted according to ASTM D 129 5 or ASTM D 1552. 8 Some specifications stipulate evaporation of sample (preceding sample decomposition) whereas other specifications call for an untreated (whole) sample. For example, Section V of the ASME Boiler Code 3 specifies ASTM D 129 5 and either ASTM D 808 6 or ion chromatography. 4 300 Liquid Penetrant Testing PART 3. Methods for Sulfur and Halogen Analysis of Liquid Penetrant Materials Detrimental Sulfur and Halogen Content of Liquid Penetrant Testing Materials There is some difference of opinion on how much sulfur or halogen content is really detrimental to any material. This is because the sulfur or halogen may be present in a number of forms and degrees of permanence and much is still unknown concerning the interactions of these elements with sensitive metals. Some specification writers feel that if the undesired elements are present in any form where they may be physically captured for analysis and in any amount over the slightest trace, the test material is unacceptable. These users require analysis of the as-received liquid penetrant testing material from its manufacturer. Other specification writers note that liquid penetrant testing materials are often sprayed on the parts and, in any case, are present only in thin layers on the part, often for considerable amounts of time. Thus, during normal testing, considerable evaporation of volatiles may take place. The detrimental sulfur or halogen content then is considered to be that of the residue after normal evaporation has taken place. This evaporation can either increase or decrease the apparent sulfur or halogen content, depending on the form in which the contaminant is present. If the contaminant is present as part of a volatile vehicle, its apparent amount is decreased by evaporation. Any contaminant present in the nonvolatile residue will be concentrated. Significance of Ionizable Compounds of Sulfur or Halogens Still other specification writers feel that there is a difference between ionizable and nonionizable sulfur or halogen. Actually, no attack on or reaction with metal can occur unless the element can ionize. Then it can leave the compound in which it was originally bound and produce new bonds with the metal. Sulfur and the halogens are normally found in both inorganic and organic forms. The inorganic forms of sulfur and the halogens usually ionize fairly easily and so are readily available for reaction with the metal. The organic forms of sulfur and the halogens usually are very tightly bound. As long as the organic compounds containing sulfur and halogen are stable, these elements may be quite harmless to metal. For this reason, some specifications require tests of aqueous extractions of the test materials. Ionizable contaminants that could attack parts are then detected whereas nonionizable contaminants are ignored. Analysis only for ionizable sulfur and halogen is probably adequate as long as the tested parts will not be subjected to conditions such as extreme heat that can break down the residual liquid penetrant testing material. The products of thermal or chemical breakdown are usually ionizable compounds that may cause attack. Sulfur Analysis of Liquid Penetrant Testing Materials The most common sulfur test specified is the ASTM D 129 5 method for total sulfur content. This sulfur analysis can be made on all organic materials. In this test, the sample is first decomposed by burning in a high pressure oxygen bomb. The sulfur present is thus converted to sulfur dioxide and sulfur trioxide. These are absorbed in a sodium carbonate solution, forming sodium sulfate. Barium chloride is added to convert the sulfur from soluble sodium sulfate to insoluble barium sulfate, which is filtered out and weighed. The amount of sulfur present is calculated from its molecular fraction in barium sulfate. This method can reliably detect sulfur contents as low as 0.1 percent (1000 gg 1 ). It becomes very unreliable below this level because of weighing inaccuracies and losses of materials that pass through the filter used to collect the final yield. Further problems can arise with many liquid penetrant testing materials containing elements such as iron, aluminum, calcium, silicon or lead or substances such as silica, asbestos or mica. All these substances cause precipitates that may falsely be measured as the barium sulfate precipitate that indicates sulfur by this test. Alternative Tests for Total Sulfur Content Several sulfur tests can be used if preferred or where ASTM D 129 5 is not applicable. This ASTM D 1552 8 test method also measures total sulfur content. It operates by burning the sample in a stream of pure oxygen to form the sulfur dioxide and sulfur trioxide. A starch iodate solution normally blue in color is bleached clear by the absorption of sulfur oxides. The titrant is often added and the color measured photoelectrically by an automatic or semiautomatic titrator. This test can measure as little as 600 gg 1 sulfur in samples on which it is applicable. 301 Liquid Penetrant System Chemistry and Effluent Waste Lamp Method for Total Sulfur Content A third method that has been proposed is the ASTM D 1266 9 lamp method for total sulfur content. Here the liquid sample is burned in a wick type lamp. An absorbent wick collects the combustion products including the sulfur oxidized to sulfuric acid. Sulfur is then either determined acidimetrically by titration against standard sodium hydroxide or gravimetrically by precipitation as barium sulfate. The lamp method is sensitive because large samples are used. It will easily find 20 gg 1 . However, it is usable only on liquids that do not contain suspended solids. Therefore, it would not work on liquid penetrant testing developers. Further acid forming elements such as phosphorus or halogens if present in substantial quantity will interfere with acidometry. Phosphorus is often a constituent of penetrant materials. Coulometric Measurement Instrumentation is available that burns the sample in a stream of oxygen and inert gas (helium or argon). This converts the sulfur to sulfur dioxide, which flows into a titration cell where it reacts with an iodine solution. An electric current is run through the cell to regenerate the iodine and the current required is a measure of the sulfur content. This test operates best with liquids but can be adapted to solids as well. The equipment for this test requires substantial investment. Bomb Turbidemetric Sulfur Test A fifth test that has been found very useful for the measurement of sulfur in materials that are completely combustible is bomb decomposition as in ASTM D 129 5 followed by turbidemetric sulfur determination as in ASTM D 516. 10 For this test, samples are burned in an oxygen bomb as in ASTM D 129 5 except that sodium hydroxide is used as the absorber instead of sodium carbonate. The bomb washings are diluted to a known volume; glycerin, sodium chloride and barium chloride are added. Light transmission measured photometrically through the mixture is proportional to the amount of barium sulfate precipitate. Ion Chromatography Still another method for sulfur testing is ion chromatography. After the sample is prepared by bomb combustion, a filtered aliquot is injected into a stream of carbonate/bicarbonate eluant and passes through a series of ion exchangers. The anions of interest are separated on the basis of their affinities for a strongly basic anion exchanger. The separated anions are measured by conductivity. They are identified on the basis of retention time as compared to standards. Quantitation is a measurement of peak area or peak height. For details of technique, see Annex A4 of ASTM E 165. 4 This method provides a single, instrumented technique for rapid, sequential measurement of common anions such as bromide, chloride, fluoride, phosphate etc. and must be considered an alternative technique for testing for those compounds as well as for sulfur. Ion chromatography is much easier and quicker to run than the standard ASTM D 129 5 and is much more sensitive with equivalent precision. Halogen Analysis of Liquid Penetrant Testing Materials Halogen is the family name for the group of elements including, from the lightest to the heaviest, fluorine, chlorine, bromine, iodine and astatine. Astatine is not stable and is not normally found in measurable or significant quantity in nature. For test purposes, it can be disregarded. Of the other four elements, fluorine is the most active and reacts differently from the remaining three. Therefore, it requires different methods of analysis than the others. In fact, fluorine is not measured as part of the total halogen analysis that is often reported. Although there is evidence that fluorine, in the form of fluoride ions, may attack some metals, only limited work has been done in this area. Limits on fluoride content are normally applied to test materials by only a few specifications, so fluoride analysis will be briefly detailed here. The reporting of total halogens (not including fluorine) is required by most specifications (including Section V of the ASME Boiler Code 3 and ANSI/ASME B 31.7 11 ) and is recommended by ASTM E 165. 4 Difficulties in Analysis of Chlorine Content Chlorine is by far the most common and possibly the most troublesome halogen. Most of the analytical procedures used for the halogens are designated as chlorine tests although, in most cases, they are sensitive to bromine and iodine as well. Thus, if the analysis must be for chlorine only, it is necessary to use special procedures to remove the unwanted bromine and iodine before the chlorine analysis is performed. Furthermore, 302 Liquid Penetrant Testing fluorine will probably not be reported because most of the tests applied in 1981 for chlorine will not respond to fluorine. The chloride ion, when present in inorganic form, is very soluble in water and so is readily available for a variety of analytical methods. Refluxing the sample is necessary to leach out all the chloride for analysis. When chlorine is present in organic form, however, it binds very tightly to the rest of the molecule and cannot usually be obtained for analysis unless the molecule is destroyed. This destruction is usually accomplished by heat, often in the presence of oxygen. The effluent gases are then absorbed in some solution for analysis. Bomb Decomposition Gravimetric Analysis of Chlorine Probably the most common test for chlorine is the bomb decomposition gravimetric analysis method of ASTM D 808. 6 For this test, a weighed sample is burned with oxygen. The gaseous combustion products are absorbed into a sodium carbonate solution, which converts the chlorides to sodium chloride. Silver nitrate is added to precipitate the chloride as silver chloride, which is then filtered out and weighed. The inadequacy of ASTM D 808 6 is generally recognized because this chlorine test is not sensitive below 0.1 percent (1000 gg 1 ) nor above 50 percent content. This test does break down and measure all organic chlorides as well as the inorganic chlorides. It measures not only chlorine but also bromine and iodine, because the silver salts of these elements are very similar to the silver chloride. The test will not measure any reportable quantity of fluorine, nor will it pick up all the bromine and iodine that may be present. Phosphorous causes interference with this test, leading to falsely high readings. Metallic Sodium Method for Chlorine Analysis Another method that was developed to measure chlorine content is the metallic sodium method. Here the sample is refluxed with metallic sodium in a mixture of butanol and toluene. The chlorides and other halogens are stripped out and form sodium salts. Methanol and acetone are added to the mixture and the chloride is titrated with silver nitrate. The end point is potentiometrically determined by using a glass electrode silver billet electrode system. This sodium refluxing method of chlorine analysis is quite sensitive, detecting as little as 3 gg 1 of chlorine. The procedure detects all the inorganic chlorine and most of the organic chlorine. Bomb Decomposition Potentiometric Analysis for Chlorine Another method for measuring chlorine is the bomb decomposition potentiometric titration method. This method decomposes the sample in an oxygen bomb as in ASTM D 808. 6 Analysis is done by potentiometric titration of the metallic sodium method described above. This method is sensitive, detecting chlorine in samples down to 3 gg 1 . The method, in common with many others, actually measures bromides and iodides along with the chlorides. This method is described in detail in Annex A2 of ASTM E 165. 4 Characteristics of Various Methods of Analysis for Chlorine Content All of the preceding methods can properly be used for the measurement of chloride ions, so long as each is used within its particular limitations. All actually measure total chloride, bromide and iodide. The bomb method of ASTM D 808 6 is less sensitive to bromide and iodide so these ions will not be fully detected. All may be subject to interference from other substances that may be present in liquid penetrant testing materials, so that test results may not always be accurate. The best way to check for accuracy is to add a known amount of chloride ion to the sample and remeasure. The percent of this known amount actually recovered will allow a good estimate of the accuracy of the analysis to be made. Ion Chromatography for Chlorine Analysis Another method for halogen testing is ion chromatography. After the sample is prepared by bomb combustion, a filtered aliquot is injected into a stream of carbonate/bicarbonate eluant and passes through a series of ion exchangers. The anions of interest are separated on the basis of their affinities for a strongly basic anion exchanger. The separated anions are measured by conductivity. They are identified on the basis of retention time as compared to standards. Quantitation is a measurement of peak area or peak height. 303 Liquid Penetrant System Chemistry and Effluent Waste Fluorine Contamination of Liquid Penetrant Materials Fluorine is the most active of the halogens, so it can cause all the problems ascribed to the other halogens plus some of its own. However, until 1981, most specifications did not test for or limit fluorine content, because adequate tests did not exist and sufficient work to determine valid limits was not done. Fluorine can exist in either organic or inorganic compounds. Only the ionizable inorganics are likely to be reactive but the organics can be converted to inorganics by the application of heat, so the safest course is to test for total fluoride rather than just readily ionizable fluoride. For this reason, it is normally necessary to decompose the test material, thus liberating the fluorides to inorganic, ionizable form. This is best accomplished by oxygen bomb decomposition. (See Annex 3 and Annex 4 of ASTM E 165. 4 ) Tests for Fluoride Content A number of tests have been used with more or less success for the determination of fluoride content. In all cases, a general precaution is in order. Fluoride tests cannot use glass apparati because the fluoride ion reacts rapidly with silica, forming a nonreactive complex resistant to all methods of analysis. This effect also makes the analysis of silica containing developers for fluoride very difficult. Possible Methods of Fluoride Analysis The fluoride ion behaves quite differently from the other halogens, so special tests are necessary to obtain valid results. Fluoride is not detected along with the other halogens by standard tests, such as ASTM D 808. 6 A number of other analytical methods have been used with varying success, however. All of the described methods are assumed to be performed using the fully ionized product of a bomb decomposition. The earliest fluoride analyses used gravimetric techniques, which are low in sensitivity, tedious and subject to many interferences. They were seldom used after 1981. Titrimetric Methods for Fluoride Analysis Many titrimetric procedures have also been developed. One of these involves titration of the dissolved fluoride with a silver nitrate potassium thiocyanate reagent and a ferric alum indicator. Unfortunately, the method is subject to many interferences, including aluminum, beryllium, iron, potassium, sodium, ammonia, phosphate and sulfate. Two other titrimetric methods operate on the basis of adding titrant and detecting the first excess of titrant, thus indicating that all fluoride has reacted. One of these titrations uses calcium nitrate titrant; the other is the lead nitratehydrochloric acid reaction. The calcium method, of course, would not work with developers that contain large amounts of calcium, and the lead nitrate method is subject to the same long list of interferences as ferric alum, noted above. Probably the most promising of the titrimetric procedures is the thorium nitrate titration with sodium alizarin sulfonate indicator. This method is only partially applicable to liquid penetrant testing materials because of certain interferences. Even in small quantities, phosphate is a serious interference, causing a red color which hides the end point completely. Phosphorous is often a constituent of test materials. Another interference is any acid insoluble solid, e.g., many developer ingredients; these completely mask the end point. Photometric Method for Fluoride Analysis A third general procedure that has been used is the reaction of the dissolved fluorides with some color forming reagent to produce a solution capable of being measured photometrically. In these tests, the depth of color is proportional to the fluoride concentration. Probably the best known of these analyses is the method of ASTM D 1179, 12 which uses a compound of sodium, 2-(parasulfophenylazo), 1,8-dihydroxy and 3,6-naphthalene disulfonate (SPADNS). This material, dissolved in water and mixed with a zirconyl chloride hydrochloric acid mixture, is bleached by fluoride. The amount of bleaching is then measured photometrically. The ASTM D 1179 method is subject to interference from aluminum, iron, phosphate and sulfate, all of which may be present in liquid penetrant materials. 12 Fluoride Electrode Method for Fluoride Analysis A device available in the 1980s eliminates most of the problems of the other methods and is now the preferred method for some fluoride measurements. This device is the fluoride specific ion electrode. When the electrode is immersed in a solution containing fluoride ions, an electrical potential is developed between the fluoride electrode and a reference electrode also placed in the solution. The potential decreases with 304 Liquid Penetrant Testing increasing fluoride content. Further, the specific ion electrode responds only to fluoride ion and hydroxide ion if hydroxide is in greater concentration than fluoride. The proportion can be controlled easily by adjusting to a lower pH. Fluoride ions can be lost through its complexing with cations such as aluminum (III) (Al 3+ ); ferric ion or iron (III) (Fe 3+ ); or silicon (IV) (Si 4+ ). This interference can be eliminated by the addition of an agent such as citrate, (cyclohexylenedinitrilo)acetic acid (CDTA) or ethylenediaminetetraacetate (EDTA). Another limitation to fluoride measurements with the specific ion electrode is that the response is not directly to concentration but to the ion activity. The activity of an ion in solution is modified by other ions in solution. If the fluoride present is ionized and the pH is properly controlled, there are essentially no interferences with this method. Ion Chromatography for Fluoride Analysis Another method for fluoride testing is ion chromatography. After the sample is prepared by bomb combustion, a filtered aliquot is injected into a stream of carbonate/bicarbonate eluant and passes through a series of ion exchangers. The anions of interest are separated on the basis of their affinities for a strongly basic anion exchanger. The separated anions are measured by conductivity. They are identified on the basis of retention time as compared to standards. Quantitation is a measurement of peak area or peak height. 305 Liquid Penetrant System Chemistry and Effluent Waste Liquid Penetrant Constituents Contributing to Waste Pollution Liquid penetrants may contain a considerable variety of chemical compounds. Among those fairly widely used are petroleum products ranging from relatively low boiling range solvents to high boiling oils. Another generic group are nonionic surface active agents. Anionic surface active agents are less commonly used because their presence in liquid penetrants has disadvantages. These are generally considered to outweigh their excellent surface active properties and, what is perhaps more important in the present context, the comparative ease with which they are biologically degraded. Other constituents present in significant proportions may be alcohols, glycol ethers and esters, to name but a few. Finally, liquid penetrants contain dyes, both visible and fluorescent. These dyes are the most troublesome contaminants because they are visibly detectable at low concentration and can be readily traced back to the source of pollution. Techniques for Reducing Liquid Penetrant Process Wastes The following chemical engineering and/or liquid penetrant processing steps are suggested to reduce the amount of liquid penetrant product residues in the waste effluent from liquid penetrant testing processing systems. 1. Minimize the amount of liquid penetrant applied. Electrostatic spray or spray application is preferable. 2. Drain the liquid penetrant as long as possible to reduce carry over. 3. Rinse postemulsifiable liquid penetrant before applying emulsifier. This liquid penetrant will separate from the rinse water and may be skimmed off or otherwise separated. This step can allow the rinse water to be recycled. 4. Apply emulsifier sparingly and allow to dwell the required time. Spray techniques use less than dip techniques of emulsifier application. 5. Rinse the part with water immediately to stop emulsification. 6. Rinse water washable or emulsified postemulsifiable liquid penetrant residues in two stages: (a) the first rinse water may be recycled until contaminated beyond use and (b) second rinse water effluent can be added to the first rinse water as makeup. 7. Rinse waters can be treated to separate out the emulsifier and liquid penetrant by using special membrane filters. Activated charcoal can be used to remove the color, oil and most of the emulsifier (wetting agent). 8. Filtration of rinse water will remove developer residues, if any. Specific recommendations for waste disposal treatment and the cost of such a system depend on numerous factors, such as (1) the type of liquid penetrant or emulsifier used, (2) whether application is by dip or spray, (3) the volume and type of work being processed, (4) the facilities and equipment in use and (5) the particular restrictions of applicable city, county, state or federal codes. Treatment Processes Applicable to Liquid Penetrant Effluents The actual effluent arising from a liquid penetrant process generally consists of a dilute emulsion formed by direct emulsification of the liquid penetrant with water, in the case of water washable liquid penetrants. Similar dilute emulsions can be formed by the action of an auxiliary emulsifier or liquid penetrant remover in the case of postemulsifiable liquid penetrants. The auxiliary emulsifying agents may contain constituents broadly similar to those of the liquid penetrants themselves and the surfactant system may be either lipophilic or hydrophilic in nature. The effluent emulsion generally contains less than one percent by weight of nonaqueous matter and is usually fairly stable. Conventional treatment can be designed to break the 306 Liquid Penetrant Testing PART 4. Techniques for Control of Pollution from Liquid Penetrant Waste emulsion and thus separate nonaqueous matter. It involves collecting and treating the effluent successively with strong electrolytes and flocculating agents in a series of vats or tanks. The coagulated contaminants are separated by filtration and finally incinerated, leaving water of acceptable purity for discharge or reuse. This is essentially a batch process, relying for its efficiency on accurate pH control. It also requires time and space: time for the various additives to become fully effective and space to accommodate the plant, the size of which is determined by the volume of effluent requiring treatment in a given period of time. Where liquid penetrant processes are used intensively and continuously, correspondingly large volumes of water are required for rinsing. More recently, membrane filtration has been used as a way of treating the effluent on a continuous basis and feeding back the purified water into the process and in effect creating a closed circuit system that is economical in the use of water. Effects of Liquid Penetrant Waste Materials in Sewage and Streams When an inspection liquid penetrant is rinsed from the test surface, it enters the rinse water as a fine, stable, oily emulsion. Typically, this emulsion is not considered to be toxic but does make the effluent turbid. It can leave an oil slick on the waters surface and even deplete some of the oxygen supply in the water. All these results lower the value of the water. Sewage regulations in some localities allow plant effluent water to contain as much as 600 gg 1 of such oils whereas other regulations allow none at all. The rinsings from liquid penetrant removal operations normally contain from 200 to 1000 gg 1 oil and are usually not an acceptable waste. Measuring Concentration of Oily Contaminants in Waste Water Oily contaminants can be detected either as hexane solubles or by the chemical oxygen demand (COD). Extraction of rinsings by hexane yields their total nonvolatile oil content. The chemical oxygen demand test measures the concentration of organic contaminants (oil, surfactants etc.) by the amount of oxygen used in oxidizing them completely by a dichromate reflux technique. The results are expressed in milligram of oxygen per liter of test solution or in parts per million. One gram of a typical oily liquid penetrant may consume about 2.5 to 3.0 g of oxygen in this manner. Thus, 1 L of rinsings containing 1 g of oily liquid penetrant may show a chemical oxygen demand of 2500 to 3000 gL 1 . More dilute rinsing may show a lower chemical oxygen demand. The chemical oxygen demand technique of monitoring pollution concentration can be related empirically to the biological oxygen demand (BOD). The biological oxygen demand test measures the oxygen used during a specified incubation period for the biochemical degradation of organic material and the oxygen used to oxidize inorganic material such as sulfides and ferrous iron. Avoidance of Environmental Water Contamination by Liquid Penetrants Where local regulations allow some oil to be present in the plant effluent, the liquid penetrant rinsings, combined with other nonoily waste streams from within the plant, may happen to yield a total effluent within the acceptable range. In most cases, though, the situation is not this simple. Then something must be done with the liquid penetrant removal operation to lower the oil content of the effluent. Worldwide, many attempts have been made to manage liquid penetrant removal without contaminating the environment. Some of these techniques use liquid penetrant compositions that meet prevailing pollution control regulations. Other control techniques alter the process so that the wastes never leave the test area. Still other techniques treat the rinsings to concentrate the oily contaminants for easy disposal. Compositions of Liquid Penetrant Materials to Avoid Pollution One approach to the pollution problem is to alter the liquid penetrants composition so that the rinsings are less objectionable. For instance, formulating a liquid penetrant to be completely water soluble would avoid the formation of turbidity and oil slicks. 13 Such rinsings might pass all tests, even though the rinse water would still be contaminated. However, complete solubility makes effluent treatment more difficult. Liquid penetrants that are completely water soluble have been available since 1953. 307 Liquid Penetrant System Chemistry and Effluent Waste Biodegradable Liquid Penetrants to Reduce Pollution Another way to alter a liquid penetrants composition is to make it biodegradable. Such a liquid penetrant need not be water soluble, yet it can have the same removal properties as oily liquid penetrants. The rinsings consist of the same sort of fine emulsion but with an important difference. Within a few days, the oily liquid penetrant is almost completely decomposed by organisms in the water. The liquid penetrant decomposition will use up a large amount of free oxygen from the stream that contains the oily wastes. This behavior limits the locations where such a liquid penetrant can be legally used. It would be objectionable if liquid penetrant effluents were emptied directly into natural bodies of water where the oxygen depletion could harm the environment. Oily waste processed in a sewage treatment plant in the presence of ample oxygen and a large population of hungry microorganisms will disappear without harmful effects. This will increase the work load of the sewage treatment plant and perhaps might result in increased sewer charges. Biodegradable types of liquid penetrant testing materials are commercially available for use in situations where suitable treatment plants are available. Recent Developments in Liquid Penetrant Waste Water Disposal The increasing regulatory restrictions on water pollution brought greater scrutiny of liquid penetrant materials getting into waste water. 14 Tests were conducted to develop a better understanding of waste water characteristics resulting from liquid penetrant materials introduced around 1990. 15 The focus was on two distinct formulation approaches compositions based on petroleum distillates and those based on surface active agents (surfactants), more commonly referred to as biodegradable liquid penetrant. The study investigated the treatability of waste water generated in the normal course of fluorescent liquid penetrant testing. The popular theory at the time was that surfactant based liquid penetrants, because they were oilfree, were more biodegradable than the traditional oil based formulations. 16 Hence making them more drain disposable. However, the study came to the conclusion that the popular theory may not be true. This very limited study found that each of the studied factors important to a publicly owned treatment works (POTW) treatment time, inhibition, respiration rate, percent biological oxygen demand (BOD) removal is more favorable with an oil based liquid penetrant than with a surfactant based liquid penetrant. It further showed that even oil free liquid penetrants test positively for oil when tested in the laboratory. Waste Water Disposal to Publicly Owned Treatment Works The greatest challenge to liquid penetrant users is knowing whether waste water from liquid penetrant processes can be sent out to publicly owned treatment works through the sewer system. 17 The first step in finding the answer is to ask the operators of publicly owned treatment works if liquid penetrant waste water is acceptable for treatment at these facilities. Waste water treatability depends on how microorganisms in waste water treatment plants respond to the waste stream. Conducting tests on fluorescent dye liquid penetrants using a respirometer simulated the treatment of liquid penetrant materials in the activated sludge system. Activated sludge consists of microorganisms cultured in a controlled environment where waste water enters and clean water (effluent) exits. The microorganisms are fed oxygen, mixed, retained for a period of time, then settled. Most waste treatment plants use the activated sludge treatment system to reduce the soluble (dissolved) organic strength waste before the waste water is discharged to a river, stream or lake. This reduction in organic strength is accomplished not only because the United States Environmental Protection Agency requires it but also because the waste water would quickly deplete the oxygen in the receiving stream. The publicly owned treatment works objective is to remove organic and solid and soluble organic waste. Anything that threatens the achievement of this goal is a concern of the facility. Drain Disposability Drain disposability is a term loosely used to denote the environmental friendliness of a particular waste when it is put in a sewer system. In the case of activated sludge systems, drain disposability would mean that the waste water can be treated without hurting the microorganisms, within the capability of the air supply system (aeration) and within the treatment time (detention time). Drain disposability largely depends on the 308 Liquid Penetrant Testing strength of the waste. Strength of waste is measured with the five day biological oxygen demand test, which measures the depletion of oxygen from primarily microbiological metabolism (respiration). The higher the biological oxygen demand, the larger the tank capacity and air requirements. Liquid penetrant materials that result in low biological oxygen demand values are more likely to be treated within the detention time allotted by publicly owned treatment works to treat waste water. In some tests, all oil based materials tested resulted in oxygen uptake rates below normal requirements whereas most surfactant based samples tested exceeded the maximum allowable range. In summary, all liquid penetrant products exerted certain amount of organic strength measured as biological oxygen demand and thus may require pretreatment before disposal to publicly owned treatment works. Techniques of Posttreatment of Liquid Penetrant Rinsings for Pollution Control Several liquid penetrant waste purification processes result in water pure enough to reuse in the rinsing step. Most of these processes rely on posttreatment of the rinsings. Posttreatment processes for liquid penetrant rinsings can be placed into four groups: (1) chemical destruction of the emulsion; (2) molecular filtration, such as reverse osmosis; (3) absorption of the emulsion onto a suitable substrate; and (4) a remover that forms an unstable emulsion that separates easily. Each of these techniques of pollution control is described in detail below. Technique for Chemical Destruction of Water Liquid Penetrant Surface active agents that allow liquid penetrant removal by water also stabilize the resulting emulsion. These surfactants can be destroyed by oxidizing agents. One of the most effective oxidizing agents is potassium permanganate. With the surface active agent destroyed, the fine emulsion can separate. The oily constituent is absorbed onto the manganese dioxide floc that forms and settles out when the permanganate reacts. This oxidation is a slow process. If permanganate is added to the collected rinsings in an amount equal to that of the liquid penetrant and the mixture is not heated, it takes about 2 h to break the emulsion. At this point, a small excess of ferrous chloride must be added to the collected rinsings to decompose any unreacted permanganate. The resulting ferric oxide floc helps absorb more liquid penetrant. The liquid penetrant floc mixture can be easily separated by filtration or decantation but the technique is slow. Pound for pound, permanganate is more expensive than liquid penetrant. Finally, the chemical additions must be painstakingly balanced so that something more obnoxious than the original liquid penetrant does not go down the sewer unneutralized. Reverse Osmosis Technique for Removal of Waste Liquid Penetrant In reverse osmosis, the impure water is forced through an osmotic membrane under pressures up to 2.80 MPa (400 lb f in. 2 ). In a single stage, from half to three fourths of the water passes through the membrane. The contaminants that cannot pass through the membrane are concentrated in the remaining waste water. Although such equipment is primarily used to separate pure water from mineral laden water, it also works well with water washable liquid penetrants that are surfactant based. Figure 7 shows schematically how a membrane separator works and includes a simple flow diagram. This is a soluble oil waste treatment system in which the separation is nearly quantitative. Beginning with municipal water laden with 2000 gg 1 of emulsified liquid penetrant (chemical oxygen demand about 4000 gg 1 ), it is possible to salvage 65 percent of the water. This recovered water contains less than 10 gg 1 impurities, largely a water soluble emulsifying agent, and had a chemical oxygen demand of 60 gg 1 . The waste water is about 35 percent of the total. This waste water contains over 90 percent of the mineral content of the original sample and all but a trace of the original liquid penetrant. Its chemical oxygen demand is around 18 000 gg 1 . The reverse osmosis process concentrates the oily contaminant into 35 percent of the original waste water, which is still too bulky for disposal. However, this concentrated waste can be forced through another osmotic membrane to salvage more pure water and shrink the volume of the waste water. Performance of Osmotic Liquid Penetrant Removal System In one series of tests with the permeator of Fig. 7, liquid penetrant rinsings were subjected to three stages of purification. 309 Liquid Penetrant System Chemistry and Effluent Waste Overall, about 94 percent of the rinsings was salvaged as pure water. The efficiency of the separation, as well as the rate of output, dropped as more concentrated water was fed to the permeator. The purified water from the first stage separation had a chemical oxygen demand of 60 gg 1 . By the third stage, enough impurities were passing to raise the chemical oxygen demand to 110 gg 1 . Table 2 shows the chemical oxygen demand values for the initial rinsings along with the values for reject and product water for each of the three successive stages. This does not represent much contamination but there is another limit on the degree of waste concentration that can be obtained. The mineral content of the waste water also increases at each stage. Eventually it exceeds the solubility and precipitates out to form scale on the membrane, which stops further action. The scale can be removed but only by shutting down the permeator and flushing it out with scale dissolving chemicals. The permeator can treat from 6 to 10 L (1.5 to 2.5 gal) of water per minute, depending on impurity level, pressure and separation ratio. Unfortunately, it is damaged by traces of chlorine and oil in water. In addition, it still requires disposal of a significant volume of waste materials. Filtration There are a variety of membrane materials and configurations. Membranes are made from polymers, stainless steel and ceramics. Membrane selection is based on the size of the particles to be separated and the chemistry of the effluent stream as it relates to the chemistry of the membrane. The standard membrane delivery system is a dead end filtration, so called because the feed fluid flow is directed at a right angle to the membrane without any attempt to control the thickness of the concentrate boundary layer at the filters surface. Rapid buildup of retained solids to the membrane surface and continuous buildup of materials severely limit the flow and separation of the effluent. The most widely used dead end filters are the cartridge filters. Generally the cartridge filters are limited to feed streams of low viscosity and solid content. Once the filter flow rate has dropped off to an unaccepted level, the cartridge filter must be discarded. To achieve relief from the accumulation of rejected effluent materials and the increasing pressure drop, filtration systems have been developed to flow the effluent stream parallel to the membranes surface in cross flow filtration, or tangential flow filtration. In tangential flow filtration, the effluent feed is pumped at a high velocity in an attempt to shear away the concentration polarization layer and minimize its effect on the separation. Turbulent flow designs 310 Liquid Penetrant Testing FIGURE 7. Schematic cross section drawing of permeator showing membrane used to pass oil free water radially outward and retain oily contaminants in tubular enclosure. Dilute oil feed Membrane Oil concentrate Oil free water Ultrafiltration process TABLE 2. Effect of successive rinses on waste water purity. Chemical Oxygen Demand (gg 1 ) ___________________________ Stage Waste Product Initial 4000 1 18000 50 2 32000 75 3 66000 110 operate at high shear rates and achieve higher filtration efficiencies. Conventional tangential flow filtration configurations include plate and frame, spiral wound (stacks of membrane sheets with turbulent promoting screens between layers), hollow fibers (bundles of filament membranes) and tubular designs (membrane tubes). The spiral wound configuration is easily serviced and is the most energy efficient tangential flow filter design. It is used in the food, chemical and environmental industries for ultrafiltration and reverse osmosis applications. Reuse of Separated Pure Water The separated pure water can be drained into the sewer or it can be reused in the liquid penetrant removal process. The small amount of impurities that passed through the first stage did not increase after six cycles of liquid penetrant reuse and contamination. Table 3 shows the chemical oxygen demand values of the product water after six cycles of reuse (see chart). Different liquid penetrants should be checked for compatibility and processing characteristics. Liquid Penetrant Removal by Adsorption Studies have shown that the clarification of effluent containing liquid penetrant waste also can be accomplished by the adsorption technique. It is essentially based on the affinity of certain absorbent particles toward typical ingredients of liquid penetrants. In practice, the oil contaminant is extracted by stirring vigorously 7 kg (15 lb) of absorbent into 10 000 L (2600 gal) of waste water containing about 0.06 percent liquid penetrant. A solution of a flocculating agent is then added and the treated batch can be separated by any of the following techniques: (1) stratification, (2) filtration or (3) centrifugation. Stratification for Recovery of Waste Water Flocculated solid particles settle loosely at the bottom of the tank, allowing the purified water to be decanted off. The bottom layer (about 20 percent of the original volume) will then have to undergo further processing such as evaporation or filtration to reduce it to a more easily disposable, compact, wet solid waste. Filtration for Recovery of Waste Water The filtration of the treated effluent batch can be accomplished by using a precoated horizontal plate filter. The end products of this technique are purified water (oil content is about 25 gg 1 ) and a wet solid cake that has only slightly more bulk than the removed liquid penetrant itself. Centrifugation for Recovery of Waste Water A successful extraction of solid particles of clay from the treated batch of effluent was also achieved by means of a simple cream separator. However, a continuous separation would require special equipment that, although industrially available, is much more expensive than an ordinary centrifuge. Water Immiscible Solvent Removers Another approach to the effluent problem is the removal of nonwater washable liquid penetrants by aqueous dispersions of volatile, water immiscible solvents. This results in an effluent whose typical composition is as follows: (1) 99 percent water, (2) 0.98 percent solvent and (3) 0.02 percent liquid penetrant. The mutual compatibility of the liquid penetrant and solvent remover and their combined immiscibility in water predetermines an easy separation of water by centrifugation or gravity stratification in a holding tank. Apart from exhibiting a slight bluish fluorescence, the recovered water is sufficiently uncontaminated (oil content less than 100 gg 1 ) to be disposed of as a regular aqueous waste. The density of the removing solvent must be either lighter than water or heavier than water in order for it to be separated by centrifugation or gravity. The considerably smaller volume of the 311 Liquid Penetrant System Chemistry and Effluent Waste TABLE 3. Purity of separated water after six cycles of reuse. Cycle Chemical Oxygen Depletion Number (gg 1 ) 1 60.8 2 57.6 3 57.2 4 64.0 5 48.0 6 64.0 remaining remover liquid penetrant mixture can then be distilled to recover the volatile solvent, leaving behind proportionally minute quantities of somewhat contaminated liquid penetrant. Or it can be skimmed off and collected for disposal as oily waste. The solvent fraction will have mostly liquid penetrant oils and traces of water and can be incinerated or used as a fuel blend. It follows that the solvent removal of liquid penetrants represents a foundation for a closed circuit system that eliminates effluent entirely by reusing water and the remover and by collecting the liquid penetrant waste in concentrated form. Technique for Liquid Penetrant Waste Purification with Activated Carbon The technique of filtration through activated carbon and other filter media for decolorizing is a well known and established technique. It is usually carried out by making a slurry of carbon in the liquid from which the color is to be removed and by heating it if possible. After a suitable reaction time, which may vary from a few minutes to several hours depending on the nature of the coloring matter and its concentration, the carbon is separated by filtration. Where large amounts of color have to be removed, it is generally necessary to repeat the process several times. It is clearly a batch process suffering from limitations of time and space. However, it has been found that it is possible to achieve the same end by allowing the effluent to flow through a bed of activated carbon and other filter media on a continuous basis. Two major factors determine the success of this technique. First, the type and particle size of carbon and filter medium is specific for a particular emulsion and apparently depends on its hydrophilic lipophilic balance (HLB). Secondly, the ratio of the rate of flow of effluent to the volume of carbon is highly critical in determining the rate at which color is removed. This in turn varies with the amount of contamination present in the carbon and filter medium, so that the rate of removal of color reduces as the degree of contamination of the filter media increases. Mechanisms of Waste Purification by Carbon Filtration Waste purification is not simply a matter of removing the organic dyes by a process of adsorption but rather a combination of this and filtration. This is confirmed by the fact that a combination of activated carbon and filter aid, such as for example, pearlite or diatomaceous earth, is more effective than either one of these on its own. If only carbon is used, the oily matter in the effluent quite soon forms an oily coating on the carbon particles that effectively prevents their functioning as adsorbents for the dye. If only pearlite or diatomaceous earth is used, only a filtration effect is attained, with little or no adsorption of dye taking place. Experimental work has shown that a sandwich arrangement, whereby the effluent flows first through the filter aid and subsequently through the activated carbon, practically doubles filter life, that is, its capacity to function before reaching actual or apparent saturation. Recycling of Purified Water after Carbon Filtration Carbon filtration, with the correct grades of carbon and filter aid, will produce clean water that may be discharged to waste. The only contaminants that may still be present in the water are the strongly hydrophilic elements of the surfactant system of the liquid penetrant or emulsifier. If the water is to be discharged to waste, it is important that only surfactants that can be readily and substantially broken down by biological organisms like biodegradable detergents be used in the liquid penetrant or emulsifier formulation. In the present state of knowledge, the choice of suitable nonionics that also meet the other required criteria is limited. However, a postemulsified liquid penetrant process in which a hydrophilic liquid penetrant remover is used presents a further problem if the rinse water is recirculated. Under such conditions the concentration of liquid penetrant remover in the rinse water gradually builds up to a level where it causes excessive foaming. This foam is deposited and dries on the parts being processed, staining them and thus hindering inspection. Adsorption of Surfactants onto Carbon Certain types of carbon are capable of removing surfactants. However, it was discovered that types of carbon that remove surfactants have no decolorizing properties with respect to the type of dye generally used in fluorescent liquid penetrants. Therefore, one has to adopt a sandwich arrangement of the two types of carbon, separated by a perforated plate, set up so that the effluent water passes 312 Liquid Penetrant Testing first through the layer of decolorizing carbon and thereafter through the layer of carbon that removes the surfactants. It was also found that if the order of the two carbon types is reversed or if they are intimately mixed or blended, the desired effect is not produced. Reverting, therefore, to what has been said previously about filter aids, a three layered sandwich consisting of one layer of filter aid and two layers of carbon is required for some applications. Disposal of the saturated carbon is best carried out by incineration, which destroys the adsorbed organic matter. If the quantities justify it, regeneration of the carbon may be economically attractive. Carbon Filters with Hydrophilic Emulsifiers In view of the relatively high price of activated carbon, the cost of the carbon filter process is not insignificant and ways of improving the economics have therefore been sought. One technique that shows worthwhile savings uses a postemulsifiable hydrophobic liquid penetrant in conjunction with a hydrophilic liquid penetrant remover. In such a system, a preliminary water rinse is carried out before immersion in the liquid penetrant remover. This rinse removes the bulk of the excess liquid penetrant by mechanical action without forming an emulsion. Then, this liquid penetrant may be recovered from the water by a centrifuge or coalescer. Both achieve almost 100 percent separation, producing clean water for reuse or discharge and actually recovering usable liquid penetrant. In one particular installation, about 4 L (1 gal) of liquid penetrants (representing about 65 percent by volume of the total dragout) are recovered daily. This process, which may be carried out continuously, achieves a double effect. It economizes in expensive liquid penetrant and, by ensuring that far less liquid penetrant is carried over to the second rinse, reduces the amount of liquid penetrant to be removed by the carbon filter, thus increasing its service life and reducing the operating costs of the process. 313 Liquid Penetrant System Chemistry and Effluent Waste Hydrophilic Concept for Recovery of Liquid Penetrant The prewash or prerinse concept, which uses a postemulsifiable liquid penetrant and a hydrophilic emulsifier, is known as the hydrophilic postemulsification liquid penetrant process. This process permits recycling of the wash water. The technique calls for removing the bulk of the postemulsifiable liquid penetrant from the surface with a plain water wash before treatment with an emulsifier. Properly formulated postemulsifiable liquid penetrant used in the prewash mode will yield nonemulsified effluent that separates by gravity when agitation ceases. The liquid penetrant should float for easy removal by skimming and for possible reuse if carefully collected. The water that remains on the bottom of the collection tank may also be reused in the prerinse step. Further processing may be required if separation is incomplete as evidenced by discoloration or organic contamination. The prewash concept can be used in closed loop liquid penetrant processes where waste water and unexpended liquid penetrants are recycled through the system rather than expelled into the environment. Prewash Concept for Recovery of Liquid Penetrant Another approach to control of liquid penetrant waste pollution uses a nonwater washable liquid penetrant and gravity separation. This system permits recycling of the liquid penetrant as well as the rinse water. The technique calls for removing the bulk of the nonwater washable liquid penetrant from the surface with a plain water wash before treatment with an emulsifier. Therefore, the principal effluent of this prewash technique is a nonemulsified mixture of liquid penetrant oil and water; the mixture quickly and completely separates by gravity. The liquid penetrant is skimmed off. The water is drawn from the holding tank and recirculated. The prewash concept can be useful in closed loop liquid penetrant processes where waste water and unexpended liquid penetrants are recycled through the system rather than expelled into the environment. Development of Prewash Concept In 1967, data were presented showing that higher liquid penetrant system performance and greater reliability were possible with the prewash approach. 18 Test data support the hypothesis that emulsifier contact time is not as critical in this system as in lipophilic postemulsification. Potential Advantages of Prewash Techniques The prewash system answers two contemporary exigencies: conservation of petroleum derived products and water pollution control. Its use results in lower material costs. Also, lower pollution abatement expenses will be experienced as the primary wash water is recycled without treatment and the second wash water is recycled with only minimal treatment. This is accomplished with a liquid penetrant system that (according to some authorities) has higher levels of performance and reliability than the existing water washable and postemulsifiable techniques. Potential Advantages of the Hydrophilic Technique The main advantage of the prewash technique from an environmental standpoint is control of water pollution. If recycling of the liquid penetrant is possible, the decrease in material costs is an added benefit. Pollution abatement expenses will decrease also as the primary wash water is recycled without treatment. Example of Prewash Separation of Liquid Penetrant Oil and Water Figure 8 compares the effluent generated by the prewash process to the effluents of the conventional water washable lipophilic and postemulsification processes. The prewash effluent shown in beaker 2 is in two layers: liquid penetrant oil and water. The liquid penetrant floats whereas the water is completely clean 314 Liquid Penetrant Testing PART 5. Recycling of Water Effluent and Postemulsifiable Liquid Penetrant except for a few globules clinging to the sides of the beaker. Effluents in the other two beakers are stable, colored emulsions. The first beaker contains the effluent of a water washable fluorescent liquid penetrant. The effluent in the third beaker is a mixture of a postemulsifiable fluorescent liquid penetrant and a lipophilic emulsifier. For photographic purposes, the effluents are 10 percent solutions, much higher than would result in practice. However, even though more concentrated than found under actual conditions, the results of stable, colored emulsions from the nonprewash processes and a floating liquid penetrant from the prewash process are accurately portrayed in the photograph. Economic Feasibility of Liquid Penetrant Waste Water Clarification In many localities, effluents from nonprewash liquid penetrant processes cannot go directly to the sewer. They require extensive treatment (1) to break the emulsion, (2) to separate the organics and (3) to clarify the water. Although water may be reclaimed after expensive processing, the separated liquid penetrant and emulsifier oils will be chemically altered and the economics of reclaiming these materials might be questioned. If the effluent is not an emulsion, then treatment costs are reduced. A nonemulsified effluent separates by gravity. Whether the pollutants float to the top of the tank or sink to the bottom depends on their specific gravity. In either case, separation takes place without expensive filtration or chemical addition. Therefore, a nonemulsified effluent can minimize water clarification cost. Equipment for Prewash Separation of Liquid Penetrant and Water The equipment required to capture and recycle liquid penetrant and rinse water from the prewash stage need not be elaborate. However, coalescers, precipitators, aerators, flocculators or similar devices are available to improve and hasten separation, especially if the water is going to be reused. In addition, there need be no chemical treatment of the effluent. The equipment needs only to use the gravity separation principle. Such equipment may involve holding tanks that allow the liquid penetrant to be skimmed from the surface of the tank or the water to be drained from the bottom. Comparison of Hydrophilic with Lipophilic Liquid Penetrant Techniques Using the hydrophilic technique is like using the lipophilic postemulsification technique inasmuch as a postemulsifiable liquid penetrant is used for both. Nevertheless, the two processes differ slightly. In the hydrophilic process, a plain water wash precedes application of an emulsifying agent. Also, instead of a full strength emulsifier solution, the hydrophilic system relies on a very dilute solution of hydrophilic emulsifier. The hydrophilic and lipophilic systems are distinguished elsewhere in this volume. Although an additional step is required to perform the hydrophilic process, this disadvantage may be more than offset by savings in the costs of material and reclaiming rinse water. Principles of Operation of Hydrophilic Liquid Penetrant Systems Prewashing with plain water physically removes from the surface all but a trace of the liquid penetrant. How completely the liquid penetrant is removed depends on factors such as (1) surface roughness, (2) water pressure and scrubbing action, (3) duration of the wash and (4) liquid penetrant characteristics such as viscosity and adhesive properties. On a typical turbine blade, for example, a nonwater washable fluorescent liquid penetrant with balanced adhesive and cohesive properties and relatively low viscosity will be removed from the surface with a water wash so that the remaining liquid penetrant is only a microscopic film, as 315 Liquid Penetrant System Chemistry and Effluent Waste FIGURE 8. Prewash liquid penetrant effluent floats (center beaker) whereas effluents from water washable (hydrophilic) liquid penetrant (left) and postemulsifiable (lipophilic) liquid penetrant (right) form emulsions. 3PT10 (287-324) 8/13/99 12:47 PM Page 315 judged from its appearance under either white light or ultraviolet radiation. Removal of Microscopic Surface Film of Liquid Penetrant after Plain Water Wash After removing all but a trace of the surface liquid penetrant with water, the next step is to convert this remaining nonwater washable liquid penetrant to a water miscible product, so it can be washed completely from the surface. This is accomplished by applying a hydrophilic emulsifier. Because the remaining liquid penetrant film is microscopically thin, the coating of hydrophilic emulsifier can also be microscopically thin. This contrasts with the conventional postemulsifiable process where a heavy layer of liquid penetrant requires a heavy layer of emulsifier. With the hydrophilic technique, only a thin coat of emulsifier is needed. The final wash effluent is produced by the thin film of liquid penetrant and overlying film of hydrophilic emulsifier. Although it is an emulsion, it is highly diluted. If it were determined that this effluent required treatment, the cost of such treatment would be greatly reduced. A charcoal filter may provide a practical means of handling the final effluent. Example of Hydrophilic Treatment of Jet Engine Turbine Blades For demonstration purposes, the white light photograph of Fig. 9 shows the three stages of the hydrophilic process on a set of relatively smooth turbine blades. Blade 1 is coated with fluorescent liquid penetrant. Blade 2 has received a prewash. Blade 3 has gone through the complete processing cycle, including treatment in a 5 percent solution of emulsifier. The procedure for removing the residual liquid penetrant after the prewash is to immerse the blade in the weak hydrophilic emulsifier solution, remove it and allow it to drain for 60 to 120 s. Then the blade is flushed with water. Allowing the emulsifier solution to drain back to the tank conserves the solution and minimizes waste water contamination by reducing dragout. Also, because the water portion of the emulsifier solution substantially evaporates during the drain, the solution concentrates for maximum effectiveness. Before using this procedure, allowing the emulsifier to sit on the part, a test should be run to determine that discontinuity detection is not degraded. Evaluation of Hydrophilic Technique on Liquid Penetrant System Monitor Panel Figure 10 is a composite ultraviolet radiation photograph that again illustrates the three stages of the process. The test piece is a liquid penetrant system monitor panel. This stainless steel panel has a chrome strip with five induced crack centers of varying magnitude whereas the balance of the panel has been sand blasted. The panel facilitates simultaneous sensitivity and washability evaluation. At 316 Liquid Penetrant Testing FIGURE 9. White light photograph shows three stages of prewash system: blade 1, liquid penetrant coated; blade 2, following plain water wash; and blade 3, after process completed. FIGURE 10. Ultraviolet fluorescent photograph shows three stages of prewash system on liquid penetrant system monitor panel: liquid penetrant coated panel (left), following plain water wash (center) and after process completed (right). the left of the composite photograph, the panel is shown after being coated with fluorescent liquid penetrant. In the middle position, the panel has been washed, substantially removing the liquid penetrant from the smooth chrome plated strip. The quantity of liquid penetrant remaining on the blasted section may be considered objectionable. The right section shows the panel after full processing; four or five crack patterns are clearly delineated and the fluorescent background on the blasted section is at an acceptable level. 317 Liquid Penetrant System Chemistry and Effluent Waste Clarification of Waste Water Containing Liquid Penetrant and Emulsifier Contaminants Modern inspection liquid penetrants are commonly made in the form of the oil phase system. This oil phase structure includes both postemulsifiable and water washable materials. The essential difference between the two types of materials is that, in the postemulsifier process, the emulsifier (detergent) is a separate process material whereas, in the water washable liquid penetrant process, the emulsifier is included as a constituent of the liquid penetrant. The liquid penetrant process causes problems of water pollution generally produced by industrial and household detergents and by solubilized oils such as industrial cutting oils and inspection liquid penetrants. A suitable clarifier material acts to precipitate detergent materials out of solution and permits separation from water by means of a conventional settling tank clarifier (described below) or a continuous flow centrifugal separator. The clarifier material completely precipitates and adsorbs dissolved detergents along with complexed oils, solvent couplers and fluorescent dyes, leaving only a few parts per million of dissolved residue. Function of Detergents in Oil Phase Liquid Penetrants The function of the detergent ingredient in oil phase liquid penetrants is to render the oil constituent of the material emulsifiable in water. The detergent acts to combine with the oily liquid penetrant and when test parts coated with the liquid penetrant/detergent mixture are washed with water, the liquid penetrant oil is flushed from the test surfaces and is emulsified and partially solubilized. The mechanics of oil emulsification imply that water added to an emulsifiable oil mixture forms molecular clusters known as micelles. These micelles become enlarged as water is added, until a point is reached where they become unstable and break apart into oil-detergent fragments that disperse in the water through dilution. Similar emulsification and solution mechanisms occur in virtually every soap and detergent composition. Hence, it is possible to group all soaps, emulsifiable oils and liquid penetrants together, insofar as their solution mechanism is concerned. Detergents Producing True Solutions of Oil in Water In some cases, where the detergent system in the emulsifier is designed for such purposes, the oil constituent of the liquid penetrant may be carried into true solution. According to the classic micelle theory, such materials remain as stable oil-water micelle clusters even with very large additions of water. Resulting mixtures often have the clarity of true solutions and behave in all respects like true solutions. When industrial process materials that contain detergents are dispersed in wash water, an unusual form of pollution occurs. Many modern detergents of nonionic nature do not respond to chemical treatment, as do some ionic solutes. Also, the water mixtures are essentially free from solid matter, except for solid soil carried into suspension by detergent action. Hence, detergent containing water wastes possess a type of pollution or contamination that is difficult to extract from the water. Pollution Resulting from Emulsifiers In the postemulsifier liquid penetrant system, contamination of wash water effluent by oil carryover on parts may be minimized by removal of excess oil before the step of emulsification by using a high pressure spray of water. The water insoluble liquid penetrant floats to the surface of the water in a sump tank and may be recovered by means of a simple weir arrangement. Although this helps in the overall problem of water pollution, it still leaves the water contamination that results from the emulsifier used in the process. Lipophilic emulsifiers are themselves mixtures of oil and detergent materials. 318 Liquid Penetrant Testing PART 6. Clarification and Distillation Recovery of Waste Water Effects of Detergent Contamination of Lakes and Streams Modern detergents, both industrial and household, have presented tremendous problems of water pollution. Several years ago, the detergent manufacturers were seeking commercially acceptable detergent products that were readily biodegradable. The idea was that all would be well if effluent waters containing such compounds would readily decompose or be eaten up by biological organisms. Unfortunately, the idea of enhancing biodegradability of detergents did not solve the pollution problem. Some lakes and streams have such high concentration of detergent contaminants that thick layers of foam or suds remain and act to choke off oxygen from marine life. In certain cases of high biodegradability, algae proliferate to the point where the algae consume all the nutrients in the water. Then the algae die and create further damage by exhausting the water of dissolved oxygen through the process of decay and decomposition called eutrophication. It is becoming more and more evident that the only permanent solution to the problem of detergent pollution of water is to remove the detergent contaminants before the waste water is discharged into outfall sewers or into waterways. Some detergents contain nonionic surfactants that are easily biodegradable. However, the issue of oxygen depletion by these surfactants has necessitated regulatory controls for waste water disposal. Possible Techniques for Clarifiers to Precipitate Detergent Wastes A possible useful approach to the extraction of detergent contaminants from water has been found in clarifier materials that have a unique property of causing the precipitation of the detergent substances out of a water solution. Although there may be some detergent materials that do not respond to the clarification action of a suitable clarifier, no such material has yet been found. The clarifier serves to completely precipitate all types of detergents, regardless of whether their chemistry has a linear or branched chain nature. In physical form, the clarifier is a fine, white, inert powder that may be readily dispersed in water. If used in the dispersion mode, the clarifier is simply stirred into the waste water being treated and is allowed to settle, as in a settling tank clarifier. Another technique of usage is the filtration mode, in which the clarifier powder is included as part or all of the filter bed. Conventional nylon fabric grids in a spin filter configuration may be used to retain the clarifier powder. After use, the clarifier powder along with precipitated detergents may be recovered by conventional backwash and flushing techniques. Chemical Behavior of Clarifier In many respects, the clarifier behaves in accordance with normal chemical principles. The clarifiers adsorption and precipitating action is a predictable function of mole ratios of the clarifier and the detergent material in the waste water. However, the reaction of the clarifier goes beyond a simple precipitation reaction with detergents. It has been found that the effect of oil solubilization in the presence of detergents is such that molecular clusters (or micelles) of oil, detergent and water occur as more or less firmly bound complexes. The result is that when the detergent substance becomes precipitated by reaction with a clarifier, it carries down with it the various other constituents that may be present in the micelle structure. This phenomenon extends even to fluorescent dyes and certain solvent couplers that may be present in an emulsifier or water washable liquid penetrant. A typical water washable liquid penetrant consists of a mineral oil, an oil-water emulsifying detergent, a solvent coupler and one or more fluorescent dyes. When clarifier is added to the water, a precipitation of the detergent takes place. In addition, many of the related constituents are carried down out of the solution along with the detergent, leaving water with only a few parts per million of residual contamination. Some pollutants in water take the form of colloidal dispersions or partial solutions of oil. Waste water of this kind may be purified in a two-step process, in which a detergent is first added to the waste water to initiate the formation of oil-detergent micelles, after which clarifier is added to the water. When the detergent is precipitated, it carries down with it the dispersed oil along with any other contaminants that can form micelle complexes with the detergent. Determination of Reaction Ratios of Clarifier If it is desired to provide an economically attractive process of water purification, it 319 Liquid Penetrant System Chemistry and Effluent Waste is necessary that the clarifier material be used at its highest possible efficiency. For rapid and complete precipitation, a slight excess of clarifier or some simple means must be used to determine when the clarifier is completely consumed. The technique of mixing clarifier into waste water and subsequent removal of the precipitate by settling or centrifuging provides a highly efficient process. The clarifier particles normally remain suspended in the water for a sufficient time to permit the precipitation reaction to go to completion. On settling or centrifuging, the detergent and other complexed materials are separated from the water, leaving only a few parts per million of dissolved material. In this mode of usage, it is desirable to determine the percent of detergent contamination of the waste water in order to add the correct quantity of clarifier. Technique of Filtering Waste Water through Clarifier In the filter mode of usage, waste water containing dissolved detergents and complexed oils is pumped through a filter bed or clarifier. In this process, the waste water passes through the filter bed rapidly, so that the precipitation reaction may not go to completion in one pass through the filter. Recirculation of the water through the filter will serve to extract the dissolved detergents fully. In this mode of usage, it is desirable to know when the filter bed of clarifier becomes exhausted. The detergency of the waste water may be determined by relatively simple techniques. One such technique involves the evaluation of surface tension of the water and comparison with samples having known detergency. Thus, it is not difficult to determine the detergent content of waste water, either before or after treatment with clarifier. Certain techniques of fluorescence analysis permit the quantitative measurement of residual contaminants by the degree of fluorescence response. Clarifier Equipment for Separating Solid Precipitates from Water Dissolved detergents are precipitated by a clarifier in the form of particulate solids. This solid material may be separated from the water by means of a settling tank clarifier or by means of a continuous flow centrifugal separator. The settling tank clarifier consists of a large circular tank with a shallow sloping bottom. Water containing suspended particles is pumped into the tank where particles settle to the bottom and clear water flows over a weir at the top. A rotating scraper moves the settled sludge in such a way that it passes out through an exit port in the bottom of the tank. This type of equipment requires a fairly large tank area and the movement of the water must be slow enough to permit settling of the fine particles of suspended matter. Centrifugal separation may be accomplished with equipment considerably smaller and somewhat more efficient in operation than the settling tank clarifier. Here the water is injected into a vertical cylindrical tank rotating at high speed. Centrifugal force drives the heavy suspended particles toward the walls of the tank. A screw conveyor or spiral scraper blade moves the separated sludge toward the bottom of the tank and out through a conical orifice. The purified water flows over a weir lip at the top of the tank and is recovered. Recovery of Clarifier Material for Reuse The clarifier material is consumed while dissolved detergent is precipitated. However, the reaction is reversible in the same way that water softening reactions are reversible. Thus, the exhausted clarifier may be reclaimed, reconstituted and reused. Onsite Clarification of Waste Water Waste water that contains detergents and oil/detergent complexes and that is discharged into sewage lines may first be treated by the clarifier process. Systems for treating water by county sanitation departments or by municipalities must be properly designed and engineered with regard to appropriate projections of flow rates. In addition, accessory systems must be included for recovery and recycling of the clarifier material and the detergents and oils that are removed from the water. For such onsite clarification, small units may be constructed from currently available components. Various bag or grid filters are available, as are various sizes of centrifugal separators. Recovery and recycling of the clarifier material itself could be carried out in separate processing plants. Closed Loop Liquid Penetrant Testing System The closed loop concept evolved from the desire to implement an apparently simple and obvious expedient of recovering the liquid penetrant process materials for reuse. However, the implementation of this idea was not all that simple. First, it 320 Liquid Penetrant Testing should be noted that complete recovery of the process materials will not be possible if any of the materials undergo a substantial physical chemical change during use. One of the reasons for failure of water purification and chemical recovery systems in the case of emulsified oils is the fact that the liquid penetrant oil undergoes a substantial physical change by being emulsified in the wash water. The micelles or molecular clusters of oil and detergent become tightly bound to water molecules and then resist separation. Slow Solubility Liquid Penetrants Some water washable fluorescent liquid penetrants, referred to as slow solubility liquid penetrants, disperse in water without undergoing emulsification. Thus, the dissolved or dispersed liquid penetrant is not tightly bound to the water in the conventional micelle emulsion structure. Another interesting feature of the slow solubility liquid penetrants is that they exhibit a high degree of discontinuity entrapment efficiency. Their entrapments in cracks are slow to dissolve and tend to remain in the cracks throughout a relatively prolonged washing period. Depletion time constants are controllable within broad limits, so the liquid penetrants can be designed to meet any desired condition of discontinuity entrapment efficiency. (Depletion time constants are described elsewhere.) Another, very important feature of the slow solubility liquid penetrants is that they exhibit very little tendency to become adsorbed onto fine porosity surfaces. Conventional emulsion forming water washable liquid penetrants and even postemulsifiable liquid penetrants, are characterized by an effect of adsorption at the liquid solid interface on test parts. Where the solid surface has a large area, as in anodized surfaces for example, this feature of adsorption produces an excessive amount of background fluorescence, resulting in poor signal-to-noise ratio. Adsorption and unwanted background are minimized in the slow solubility liquid penetrants. The first of the two new chemical categories is exemplified by liquid penetrants characterized by a relatively rapid rate of wash removal with a hot water spray wash at temperatures in the vicinity of 55 C (130 F). The second category liquid penetrants are chemically different and provide progressively larger indication depletion time constants. Recycling Liquid Penetrant after Solvent Distillation The new slow solubility liquid penetrants may be recycled in a closed loop operation through three stages. First, surface liquid penetrant is stripped from parts by means of a pressure spray of wash water. The liquid penetrant thus removed does not dissolve rapidly in the water; instead, it tends to float on the surface of the wash water and may be drawn off over a drainage weir and recovered. Recycling of Wash Water The wash water is circulated from a reservoir through wash nozzles and back to the reservoir. After considerable use, the wash water becomes saturated with dissolved liquid penetrant. It is necessary to extract the dissolved liquid penetrant continuously from the wash water to preserve its ability to dissolve liquid penetrant from test surfaces. This extraction is carried out as outlined above. If it were required to purify the used prewash water enough to drink, liquid penetrant contamination would have to be reduced to a value of about 5 gg 1 or less. For purposes of prewash treatment in the closed loop process, it is not necessary for wash water to be so pure. Process Diagrams for Closed Loop System For reasons of economic feasability, the closed loop, water washable liquid penetrant testing technique has not been developed commercially, specified in standards or implemented by industry as of 1999. The process would consist essentially of five interlocking loops (Fig. 11). 321 Liquid Penetrant System Chemistry and Effluent Waste FIGURE 11. Interlocking loops of closed loop system for water washable liquid penetrant testing process. Parts Process Skim Water Liquid penetrant Solvent Distillation Solvent extraction Inspect Liquid penetrant 1. Test parts would be processed through the steps of (a) liquid penetrant application, (b) wash removal of surface liquid penetrant and (c) inspection for indications. Development, of course, would be included as part of the inspection step. 2. Part of the liquid penetrant would float and would be skimmed and retained for testing to determine its suitability for continued use. 3. Used wash water would be cycled through a solvent extraction stage and the purified water would be returned to the wash water reservoir. 4. The solvent that contains extracted dissolved liquid penetrant would be cycled through a distillation column and recovered solvent would be returned to the extraction column. 5. Liquid penetrant recovered by distillation of extraction solvent would be retained for testing to determine its suitability for continued use. 322 Liquid Penetrant Testing 1. Spanner, J.C. [Sr.] Methods and Reasons for Measuring the Chloride Content in Liquid Penetrant Materials. Materials Evaluation. Vol. 30, No. 6. Columbus, OH: American Society for Nondestructive Testing (June 1972): p 126-135. 2. ASTM A 380, Recommended Practice for Cleaning, Descaling, and Passivation of Stainless Steel Parts, Equipment, and Systems. West Conshohocken, PA: American Society for Testing and Materials (1996). 3. ASME Boiler and Pressure Vessel Code. New York, NY: American Society of Mechanical Engineers. 4. ASTM E 165, Standard Test Method for Liquid Penetrant Examination. West Conshohocken, PA: American Society for Testing and Materials (1995). 5. ASTM D 129, Standard Test Method for Sulfur in Petroleum Products (General Bomb Method). West Conshohocken, PA: American Society for Testing and Materials (1995). 6. ASTM D 808, Standard Test Method for Chlorine in New and Used Petroleum Products (Bomb Method). West Conshohocken, PA: American Society for Testing and Materials (1995). 7. ASTM D 2441, Standard Test Method for Hydrolyzable Chlorine Compounds in Chlorinated Aromatic Hydrocarbons (Askarels) by Refluxing. West Conshohocken, PA: American Society for Testing and Materials (1995). 8. ASTM D 1552, Standard Test Method for Sulfur in Petroleum Products High-Temperature Method). West Conshohocken, PA: American Society for Testing and Materials (1995). 9. ASTM D 1266, Standard Test Method for Sulfur in Petroleum Products (Lamp Method. West Conshohocken, PA: American Society for Testing and Materials (1991). 10. ASTM D 516-90, Standard Test Method for Sulfate Ion in Water. West Conshohocken, PA: American Society for Testing and Materials (1995). 11. ANSI/ASME B 31, Code for Pressure Piping. Washington, DC and New York, NY: American National Standards Institute. 12. ASTM D 1179, Standard Test Methods for Fluoride Ion in Water. West Conshohocken, PA: American Society for Testing and Materials (1993). 13. Goff, R. and S. J. Robinson. Water-Base (WB) Penetrants Advantages and Disadvantages. ASNT Fall Conference and Quality Testing Show Paper Summaries [Nashville, TN]. Columbus, OH: American Society for Nondestructive Testing (October 1998): p 114-116. 14. Robinson, S.J. Issues Concerning the Disposal of Waste Penetrant Materials (Back to Basics). Materials Evaluation. Vol. 49, No. 8. Columbus, OH: American Society for Nondestructive Testing (August 1991): p 962-967, 969. 15. Holmgren, V. and M. Plamoottil. Testing of Detrimental Elements in Penetrant Materials. 1992 ASNT Fall Conference and Quality Testing Show [Chicago, IL]. Columbus, OH: American Society for Nondestructive Testing (November 1992): p 179-180. 16. Holmgren, V. Penetrant Materials Are They Biodegradable? 1989 Fall Conference ASNT Program and Paper Summaries [Valley Forge, PA]. Columbus, OH: American Society for Nondestructive Testing (October 1989): p 39-40. 17. Hessinger, P. and M.L. White. Treatment Alternatives for Liquid Penetrant Rinse Water. Materials Evaluation. Vol. 56, No. 8. Columbus, OH: American Society for Nondestructive Testing (August 1998): p 969-970. 18. Birley, R.E., N.H. Hyam and T. Tebbenham. Removal Techniques in Liquid Penetrant Inspection Processes, Their Development and Effect on Sensitivity. Proceedings of the Fifth International Conference on Nondestructive Testing [Montreal, Canada, May 1967]. Ottawa, Canada: Queens Printer (1969): p 222-225. 323 Liquid Penetrant System Chemistry and Effluent Waste References