Avoiding Aluminum Nitride Embrittlement in Steel Castings for Valve Components

V-Rep 84-1

Edward Valves

Avoiding Aluminum Nitride Embrittlement in Steel Castings for Valve Components

By William C. Banks, Senior Metallurgical Engineer Materials Engineering Department Flow Control Division, Rockwell International First Published 1984

The Flow Control Division of Rockwell is concerned that low ductility steel castings, embrittled by aluminum nitride, may not be detected prior to their becoming valve components. This concern stems from the trend in recent years away from green sand (clay-bonded) molding in favor of airset sand molding (using organic binders), notable effect of the latter being to produce castings that cool more slowly through the temperature range where aluminum nitride precipitates around the primary austenite grains of the steel. The lack of provisions in the ASTM specifications to forestall the shipping of castings embrittled by aluminum nitride has prompted Rockwell to spearhead a task force with the objective of removing this deficiency in the specifications. The accompanying article discusses the present status of activities both within Rockwell and in ASTM aimed at ensuring that steel valve body and related pressure-containing castings do not contain deleterious amounts of aluminum nitride, regardless of the foundry practices used to produce them. Steel castings are the mainstay of the steel valve manufacturing industry. They are a highly cost-effective means of obtaining medium- to large-size valve bodies of the desired shape, wall thickness, and composition in a broad range of valve sizes and types. Over the years, the steel foundry industry has developed modifications in steel melting practices, molding methods, welding processes, and nondestructive testing to keep pace with the ever-increasing concern for quality standards shown by the

users of steel castings. Some of these modifications introduce new manufacturing variables that can affect the metallurgical properties and internal quality of the castings; therefore, modifications in practice and control procedures may become necessary. A case in point is a trend in recent years for foundries to discontinue their use of green sand (clay-bonded) molding in favor of air-setting sands bonded with organic binders. In addition to increasing productivity, the air-set process enables the foundries to make cleaner, better dimensioned castings with more faithful reproduction of pattern details. Although quantitative data is not available, it has been observed that castings cool much slower in the organically bonded sand molds than they do in clay-bonded green sand molds. The difference in cooling rate in the air-set molds is generally attributed to the mild exothermic mold/metal reaction from the organic binders plus the insulating effect of the dead-air spaces created when the organic binders are volatilized by the beat from the casting. Furthermore, on the basis of such studies as reported by T. T. Rick* , molding mixtures with the greatest apparent density will have the greatest heat abstracting capacity. By virtue of the different methods used to compact clay-bonded sand and organically bonded sand against the patterns, green sand rams to a somewhat greater density than air-set sand and therefore exhibits a greater capacity for heat abstraction which in turn results in faster cooling of the casting.

FIGURE 1: Fracture surface of section from buttweld end of a Grade WCB carbon steel valve body casting exhibiting a rock candy appearance, characteristic of aluminum nitride embrittlement.

Introduction

FIGURE 2: Fine-grain fracture appearance typical of normal cast carbon steel (Grade WCB). Section is from a 1.5 -inch (37 mm) thick test block representative of a heavy section valve body.

*Heat Abstraction by Molding Materials, Foundry. 77 1949, p. 96-97, 182, 184, 186,188.

Avoiding Aluminum Nitride Embrittlement in Steel Castings for Valve Components

Slow cooling rates, in themselves, are normally not considered detrimental in the production of steel castings. However, the common use of aluminum as a deoxidizer in the molten steel can, under certain circumstances, when combined with a slow cooling rate and the presence of nitrogen, reduce the ductility and toughness of the steel by a precipitation phenomenon. When this embrittling condition is present, fracture appearance is characterized by coarse angular facets and the phenomenon is referred to as intergranular or rockcandy fracture. Figure 1 shows the typical rock-candy fracture exhibited by an embrittled weld-end section from a carbon steel (Grade WCB) valve body. For comparison, the normal fine-grain fracture of this grade of steel is illustrated in Figure 2. Experience indicates that under certain welding conditions, cracking may be encountered in the heat-affected zone of castings susceptible to intergranular fracture. Because the cooling rates of castings produced by modern air-set molding procedures are slower, such castings are likely to be more sensitive to this problem than similar castings produced by older green sand methods. Many years ago it was established that the compound known as aluminum nitride (AIN) could precipitate in the form of thin films on the boundary surfaces of the primary austenite grains and be a cause of intergranular fracture. The primary variables controlling the susceptibility of a steel to this type of embrittlement are the aluminum content, nitrogen content, and the cooling rate through the austenite region of
1800 A: B: C: D: E: 0.127% 0.09% 0.06% 0.03% 0.01% B AI AI AI AI AI 3272

1600

CURVE Y CURVE Z

0.018% N

2912

1400 TEMPERATURE IN C CURVE X 1200

2552 TEMPERATURE IN F

2192

1000

1832

800 CURVE X1" SECTION COOLING IN GREEN SAND CURVE Y212" SECTION COOLING IN GREEN SAND CURVE Z212 SECTION COOLING IN AIR-SET SAND 600

1472

1112

400

101

102
(1 MIN. 40 SEC.)

103
(16 MIN. 40 SEC.)

104 TIME T IN SECONDS

(2.78 HR.)

105
(27.78 HR.)

106
(277.78 HR.)

107

752

(10 SEC.)

(2777.78 HR.)

FIGURE 3: Schematic representation of the interrelation of aluminum content, nitrogen content, and cooling rate on the formation of aluminum nitride in cast steel. lntersection of a cooling curve with any of the curves (A, B, C, etc.) representing the start of AIN precipitation results in the presence of AIN in the steel at room temperature.

the iron-carbon diagram. In the past, this problem was manifested only in very heavy wall castings produced in green sand molds. However, long-established melting and deoxidation practices proven satisfactory for steel cast in green sand molds may not necessarily provide assurance of avoiding aluminum nitride embrittlement in steel cast in air-set molds. The slower cooling rate in air-set molds intro-

duces a new variable. The purpose of this article Is to describe some technical activities aimed at ensuring that steel valve body and related pressure-containing castings are free of deleterious amounts of aluminum nitride, regardless of the foundry practices used to produce them.

Avoiding Aluminum Nitride Embrittlement in Steel Castings for Valve Components

Factors Affecting the Formation of Aluminum Nitride
Although the amount of aluminum added as a deoxidizer to the steel can normally be closely controlled, the amount recovered and reported as a residual varies considerably, This variation is largely due to the different states of oxidation the steel is in at the time of the aluminum addition and also due to physical mixing problems associated with a lightweight, relatively low-melting-point, reactive metal being added to molten steel. thickest section in any given heat of steel he pours will spend a minimum amount of time in the temperature range where the aluminum nitride precipitation is occurring, For a given nitrogen Content, it is important that the residual aluminum content be decreased as the section thickness being cast increases, The same principle applies if the cooling rate of a given casting is decreased as a result of making the casting in an air-set mold. In other words, the critical thickness for avoiding aluminum nitride formation with a given aluminum and nitrogen content in the steel decreases as the cooling rate decreases. Thus, for a given thickness, aluminum levels must be reduced and the nitrogen content carefully controlled when air-set molding practices are employed, In view of the interacting factors contributing to the formation of aluminum nitride, evaluation methods are needed to determine whether conditions were such that harmful amounts of aluminum nitride may have formed in a given casting. The problem primarily encompasses castings with thick cross sections and those of a compact design having a relatively low ratio of surface area to volume. In the latter case, such castings cool slowly because less area is available to effect heat transfer to the mold. It is fortunate that a relatively wide range of casting conditions exist wherein aluminum nitride embrittlement is not a problem. In his paper presented to

Methods for Detecting Aluminum Nitride Embrittled Castings

Because most steel foundries use electric arc melting, the nature of this process can introduce unknown amounts of nitrogen into the steel, and despite efforts to flush it out, maintaining consistently low levels is difficult. An additional problem with nitrogen control is that the analytical equipment for nitrogen determinations is costly and the time required for analysis may be too long for the melting departments needs. Few foundries have in-house capability for making nitrogen analyzes. Reliance must therefore be placed on standardizing melting practices as much as possible by using techniques associated with the attainment of low nitrogen levels as determined subsequent to the casting of the steel. Valve body castings are produced in a wide variety of shapes and sizes, so the foundry must contend with a considerable range of section thicknesses and consequently a wide range of cooling rates. As the schematic graph of Figure 3 illustrates, the foundryman must control his ALUMINUM ) and nitrogen levels so that the

FIGURE 4: Reference photographs of macroetched cast steel showing 10 levels of severity of intergranular network structures indicative of the presence of aluminum nitride precipitation in the primary austenitic grain boundaries.

Avoiding Aluminum Nitride Embrittlement in Steel Castings for Valve Components

the 17th Annual (British) Steel Casting Research and Trade Association (SCRATA) Conference in 1971, H. M. Kuhn* reported that in unalloyed steel castings with wall thicknesses less than 4 inches (100 mm), restricting the aluminum content to 0.08% maximum prevents the occurrence of intergranular fracture. Presumably this conclusion was based on experience with castings made in clay-bonded green sand molds because the newer organically bonded molding sands were not yet in general use at that time. As part of its overall quality assurance program, the Rockwell Edward Valve organization (in the early 1950s) began evaluating the steel in its large valve bodies by conducting metallurgical tests on specimens from lugs or bars integrally cast on the bodies. For this purpose, a 2 inch by 2 inch by 6 inch long (50x50x150 mm) bar was used on all body castings with a 1.5 inch (37 mm) thick wall or greater in combination with a weight of more than 1000 lb (455 kg). This extension bar was enlarged to 3 inches by 3 inches by 6 inches (75x75x150 mm) on castings with a wall thickness of 4 inches (100 mm) and greater. This testing was done in addition to that required by the basic ASTM specifications (such as A216 and A217) for pressure-containing steel castings. Evaluation of the steel in accordance with these specifications requires, as a minimum, the tensile testing of specimens from ASTM A370 separately cast keel blocks whose coupon cross-sections are 1 inch by 1.25 inch (25 by 32 mm). The Rockwell Edward program thus provided the opportunity to gain data on the tensile properties of the cost steel In the heavy sections of valve bodies through testing of specimens that were representative of the actual products.
TABLE 1 Descriptive Data Applicable to Network Structures Shown In Figure 4 Rating 1 2 3 4 5 6 7 8 9 10 Delineation Width Fine .001 Fine .001 Fine .001 Fine .002 Fine .002 Medium .005 Heavy .010 0.020 1/32 1/16 % Boundary Outline 20 40 60 50 100 100 100 100 100 100

Note: These ratings are based on the physical width and continuity of the precipitate pattern developed by the acid etchant on the primary austenitic grain boundaries of the cast steel. A 0 rating indicates that no delineation of the grain boundaries has been observed. Supplementary testing is normally conducted to determine the final disposition of castings with ratings of 5 or greater.

of aluminum nitride, Etching of smoothground cross sections of these heavy section coupons in a hot (160180F) (7182C) 1:1 hydrochloric acid solution revealed significantly different macroetched structures depending upon the amount of aluminum nitride precipitation that was present in the primary austenitic grain boundaries. From the accumulated body of data it was possible to establish a series of reference photographs of the etched steel indicating various degrees or severity levels of network (chicken-wire) structures created by increasing widths of the grain boundaries and greater degrees of their continuity. The graded series of these reference photographs are illustrated in Figure 4, and Table 1 is used in association with the photographs to give additional data for establishing a network rating for a given steel specimen. Analysis of network ratings for carbon steel (Grade WCB) indicated that severity ratings below a 3 had a negligible effect on tensile ductility values whereas ratings of 3 or 4 had a statistically noticeable effect on these values. However, ratings of 4 or less were generally considered acceptable. To illustrate how deleterious an aluminum nitride network rating of 8 can be, the data in the following table are presented. Specimens 1 and 2 were from a 2.5 inch (62 mm)-thick buttweld end section of Grade WCB steel similar to the one shown in Figure 1. The steel had residual aluminum and nitrogen levels of 0.10% and 0.012%, respectively, and exhibited a very obvious rock-candy fracture with a network rating of 8. 5

In the course of this years long program involving close cooperation with the steel foundries supplying the castings, it was determined that a major cause of deficient levels of ductility in the cast-on coupons (as indicated by low values of percent elongation and percent reduction of area in the tensile test) was poor melting and deoxidation practices that promoted the formation

*Electric-Arc Furnace Steelmaking Practice With Regard to Special Foundry Problems, proceedings of the above SCRATA conference, Paper No. 2 or Foundry Trade Journal 1972 Vol. 132 (2876) 81-88.

Avoiding Aluminum Nitride Embrittlement in Steel Castings for Valve Components

Specimen No. 1 2 Foundry Keel Block from Heat ASTM A216 GradeWCB Requirements, min. Tensile Strength, ksi (MPa) 75.2 75.5 76.2 (518) (520) (528) Yield Strength, ksi (MPa) 63.6 71.5 42.8 (438) (493) (295) Elongation in 2 (50 mm) % 11.5 9.5 28.0 Reduction of Area % 14.1 10.8 51.0

70.0

(485)

36.0

(250)

22.0

35.0

It is interesting to note that the minimum tensile and yield strength requirements for Grade WCB steel are met by the aluminum-nitride-embrittled steel despite Its exhibited sharply lower ductility values in the relatively thick weld end section. Such results are typical of steel embrittled by aluminum nitride. No clue as to the tendency of the heat toward aluminum nitride precipitation is observable from the results of the keel block specimen. Should castings with harmful amounts of aluminum nitride precipitates fail to be detected and enter the field, the concerns will be the same as those for any product depending upon a steel with an expected level of ductility but which it may not have. The ability of a steel to accommodate accidental overloads in components with stress concentration areas or sudden impact loads is certainly decreased when the inherent ductility is impaired by the presence of embrittling, intergranular films. As mentioned earlier, evidence exists that such films can contribute to the formation of cracks in the heat-affected zone of highly restrained welds even in a carbon steel.

Despite the fact that it has been known for over 30 years that aluminum-nitride-embrittled steel castings could be produced, the ASTM specifications for steel castings have been strangely silent on the subject. Evidently field failures attributable to this embrittlement phenomenon have been relatively infrequent and the difficulty of obtaining a consensus among the foundries as to what specific control measures would be effective have inhibited initiatives toward developing an ASTM standard practice, Just as in the Rockwell Edward approach, other responsible companies dependent upon reliable steel castings have protected themselves through appropriate provisions for AIN control in private-type material purchase specifications. Although Rockwell has encountered only a few isolated instances of aluminum-nitrideembrittled steel valve bodies, the Flow Control Division believed that the potential hazard for its type of productsespecially those used in gas transmission pipelines and steam power systems-merited bringing

Provision for Aluminum Nitride Control In ASTM Specifications

the issue to the attention of ASTM Subcommittee A01. 18 on Steel Castings. In this way the entire steel valve manufacturing industry, as well as all users of steel castings for pressure containment, in particular, would be alerted to the hazard and available for supporting ASTM specification action. As a result of the Rockwell initiative, a task group was formed in May, 1982, under the chairmanship of the author to develop appropriate wording for insertion in ASTM Specification A703 General Requirements Applicable to Steel Castings for Pressure-Containing Parts. Initially it was proposed that the aluminum nitride control provision should be made a part of the main body of A703 and thereby automatically apply to all the specifications for pressure-containing castings linked to A703 namely, A216, A217, A351, A352, A389, A487, and A643. However, because these specifications are widely used for purchasing both pressure-containing and nonpressure-containing castings, many of which have small cross sections and are relatively lightweight, adding a mandatory provision for aluminum nitride control across the board would place an unjustifiable economic burden on the steel casting industry. Therefore, the task group decided to make its proposal as a Supplementary Requirement in A703 for use by anyone concerned about possible aluminum nitride embrittlement in his steel castings. Although at this writing precise wording for the control provisions under the title Macroetch Test has been approved by ASTM Subcommittee A01.18, and is to be 6

Avoiding Aluminum Nitride Embrittlement in Steel Castings for Valve Components

balloted in the Main Committee A01 during 1984 it would be improper to include the balloted item in this technical article at this time, However, the key features of the proposal include: (1) The total residual aluminum content of the heat is to be reported to the purchaser (2) When the aluminum content exceeds 0.08%, the foundry is to conduct a macroetch test on a sample representative of the heaviest section of a casting in the heat by the procedure given in ASTM E340. (3) The resulting etched sample is to be compared and rated with the reference photographs similar to those shown in Figure 4 of this article. If the etched network rating does not exceed Level 4, the heat is acceptable. (4) Provision is made for the disposition of castings with etched ratings above Level 4 by using supplementary testing, for example, fracture examination, bend tests, and tensile tests, to confirm the actual degree of embrittlement. High-temperature solution annealing for possible salvaging of rejected castings is to be permitted.

In the interim, until the ASTM steel casting specifications are revised, the internal Rockwell Material Specifications used by its Flow Control Division plants contain aluminum nitride control provisions based on the proposals now moving through the ASTM approval process. The Rockwell specifications contain one additional precaution not yet included in the ASTM proposal but which will be suggested as a further revision once the basic control provision is in place in A703. This further control feature requires the checking of socalled heavy wall castings defined as having section thicknesses exceeding 1.5 inches (37 mm) and a weight of at least 1000 lb (455 kg) by the macroetch test when the residual aluminum content of the heat exceeds 0.06%. As an alternative to the macroetch test for these heavy-wall castings, tensile testing of a coupon removed from a heavy section of the casting may be conducted after heat treatment, the results of which are required to meet the minimum valves specified by ASTM for the grade of steel involved. It is the intent of this article to alert the users of steel castings, particularly those in the valve industry, to the possibility that an old but troublesome phenomenon, aluminum nitride embrittlement, may occur with greater frequency as a result of foundries replacing the old clay-bonded green sand molding methods with new

Provision for Aluminum Nitride Control In Flow Control Division Specifications

organically bonded (air-set) molding methods. This concern is based on the observation that castings made with aluminumdeoxidized steel in air-set molds cool more slowly than those made in green sand molds; therefore, they may be more prone to the formation of thin, embrittling films of aluminum nitride around the primary austenite grains. Such films markedly reduce the ductility of the steel as measured in the conventional tension tests. Since the critical thickness for avoiding aluminum nitride with a given aluminum and nitrogen content in the steel decreases as the cooling rate decreases, the significance of this change in the molding method must be recognized. A level of residual aluminum of 0.08% max. in a heat is proposed, above which various evaluation methods are employed to detect any possible harmful level of aluminum nitride embrittlement. Rockwell has initiated actions through its participation in ASTM committee activities to focus national attention on the aluminum nitride threat to the integrity of valve bodies and other related pressure-containing castings. Rather than wait for the provisions for aluminum nitride control to clear the ASTM approval process, Rockwell has implemented its own Flow Control Division material specifications to ensure that currently supplied cast steel valve bodies are not embrittled by aluminum nitride, regardless of the molding methods used to produce them.

Summary

Flow Control Division

Edward Valves

Flowserve Corporation has established industry leadership in the design and manufacture of its products. When properly selected, this Flowserve product is designed to perform its intended function safely during its useful life. However, the purchaser or user of Flowserve products should be aware that Flowserve products might be used in numerous applications under a wide variety of industrial service conditions. Although Flowserve can (and often does) provide general guidelines, it cannot provide specific data and warnings for all possible applications. The purchaser/user must therefore assume the ultimate responsibility for the proper sizing and selection, installation, operation, and maintenance of Flowserve products. The purchaser/user should read and understand the Installation Operation Maintenance (IOM) instructions included with the product, and train its employees and contractors in the safe use of Flowserve products in connection with the specific application. While the information and specifications contained in this literature are believed to be accurate, they are supplied for informative purposes only and should not be considered certified or as a guarantee of satisfactory results by reliance thereon. Nothing contained herein is to be construed as a warranty or guarantee, express or implied, regarding any matter with respect to this product. Because Flowserve is continually improving and upgrading its product design, the specifications, dimensions and information contained herein are subject to change without notice. Should any question arise concerning these provisions, the purchaser/user should contact Flowserve Corporation at any one of its worldwide operations or offices.

For more information about Flowserve Corporation, contact www.flowserve.com or call USA 1-800-225-6989. FLOWSERVE CORPORATION FLOW CONTROL DIVISION Edward Valves 1900 South Saunders Street Raleigh, NC 27603 USA Toll- Free Telephone Service (U. S. and Canada) Day: 1-800-225-6989 After Hours Customer Service 1-800-543-3927 US Sales Offices Phone: 919-832-0525 Facsimile: 919-831-3369 Facsimile: 919-831-3376 Visit Our Website www.edwardvalves.com

2003 Flowserve Corporation, Irving, Texas, USA. Flowserve and Edward Valves are registered trademarks of Flowserve Corporation.

V-Rep 84-1 3/03 Printed in USA