QUARTERLY Volume 6, Number 3

Mobile Parts Hospital Making Replacement Parts in the Field MaterialEASE: Tips and Pitfalls of Corrosion Testing

AMPTIAC is a DOD Information Analysis Center Administered by the Defense Information Systems Agency, Defense Technical Information Center and Operated by IIT Research Institute

Editorial: Survival of the Fittest Every Defense researcher knows that our ultimate responsibility is to help our soldiers and sailors by providing them with the best possible equipment and technology available. We take this mission very seriously, and realize that our efforts will eventually save the lives of our countrymen by enabling them to do their jobs better, from farther away, with more precision and accuracy. Our military is in the midst of a fundamental change in the type of war it will wage in the future. We have spent the better part of the past fifty years preparing to fight a well-organized enemy, with massive war-making tools on the land and in the sea and air. This enemy played by a similar rulebook to ours, structured itself similarly, and basically had a fighting ability and sense of morality not unlike our own. Our experiences with enemies in Europe and Japan taught us much about how to fight, but in the end we always knew who our enemies were, and could invite them to a negotiating table to discuss peace. In the 60s and 70s, we learned a valuable lesson: Much of the technology we had developed proved less effective against a loosely organized enemy with basic weapons and an intense desire to defend their homeland. We learned to fight a guerrilla war, in the jungles, ditches and swamps, with foot soldiers on the ground the main asset to face the enemy. We learned a lesson we had initially taught the British almost two centuries before: loosely organized militia with plenty of hiding places will pose a difficult challenge to organized and structured troops studying a well-defined rule book for how wars are supposed to be fought. We also learned that intelligence gathering was going to be a key ingredient of future combat. Our experiences in the Far East thirty years ago again taught us much, but our primary enemy was still the Soviet Union and Southeast Asia was only one battle against Communism, not the entire war. I am not a military historian, and these views are simply my take on recent history that has led us to this point in time. Our

former enemies are now our allies, our new enemies are shrouded in mystery and our technology has many difficulties in breaching their organizations. Our military faces far greater challenges in today’s world: How do you fight an enemy that we know so little about? How do you find out more? How do you fight him in his own land, streets, and buildings? How do you shift your war-making tools to policing tools after the battles are won? Indeed, how do you find someone who has practiced hiding to the point that it is an innate skill? Our military is carrying out its new mission, but it is a mission that is still evolving. We are learning how to use the tools we have in new ways, to accomplish new tasks that weren’t envisioned just a few years ago. We are learning to focus more on information in the battlespace, such that resources are used in the most effective manner possible. Constant vigilance is our only ally, because our new enemy is very creative, devious and, above all, patient But back to us, the Defense materials and processes community. Our mission of supporting our military is a challenging one. We are often so far from the "point of the spear" that we risk loosing sight of our mission, and sometimes we even struggle with our own political opinions. Simply put, our job is to create the new materials and technologies needed for remote sensing, force protection, intelligence gathering, guidance, control, and weapons systems. Existing tools will be modified and new ones designed to fit the needs of a new and very different kind of war. Advanced materials are just as important in enabling today’s new systems and capabilities as they were in the past. We must also realize that our focus may have to evolve to recognize new opportunities. But when all is said and done, our mission is simply to provide our warfighters with the equipment they need to do the job, and in that, we will not fail. Wade G. Babcock Editor-in-Chief

CORRECTION: On page 7 of the last issue of AMPTIAC Quarterly, in Figures 3 and 4 of the article “Lowering the Cost of Titanium,” the Internet address for AeroMet Corporation was incorrectly identified. The correct Web address is www.aerometcorp.com. We apologize for any inconvenience this may have caused.

Editor-in-Chief Wade G. Babcock Creative Director Cynthia Long Information Processing Judy E. Tallarino Patricia McQuinn Inquiry Services David J. Brumbaugh Product Sales Gina Nash Training Coordinator Christian E. Grethlein, P.E.

The AMPTIAC Quarterly is published by the Advanced Materials and Processes Technology Information Analysis Center (AMPTIAC). AMPTIAC is a DOD sponsored Information Analysis Center, operated by IIT Research Institute and administratively managed by the Defense Information Systems Agency (DISA), Defense Technical Information Center (DTIC). The AMPTIAC Quarterly is distributed to more than 25,000 materials professionals around the world. Inquiries about AMPTIAC capabilities, products and services may be addressed to David H. Rose Director, AMPTIAC 315-339-7023 EMAIL: URL:

a m p t i a c @ i i t r i . o rg

h t t p :/ / a m p t i a c . i i t r i . o rg

We welcome your input! To submit your related articles, photos, notices, or ideas for future issues, please contact: AMPTIAC ATTN: WADE G. BABCOCK 201 Mill Street Rome, New York 13440 PHONE: 315.339.7008 FA X : 3 1 5 . 3 3 9 . 7 1 0 7 EMAIL:

a m p t i a c _ n ews @ i i t r i . o rg

Stan Kolisch IIT Research Institute

Materials are enablers of technology – this is an axiom that most materials specialists know and embrace. Typically, we tend to think of such materials in terms of fielded hardware and the added performance they provide. What is not always considered yet equally true, is that materials enable new and improved manufacturing processes and technologies as well. Thus, new materials not only afford us with systems that perform better, but they also allow us to produce such systems at higher rates, lower costs, and with higher degrees of quality and reliability. The advances in powder metallurgy over the past two decades have facilitated a series of rapid prototyping and manufacturing technologies, which are allowing us to redefine the conventions of manufacturing and production. The Army’s Mobile Parts Hospital is an excellent example of harnessing the potential of such emerging technologies. - Editor Introduction The U.S. Army, through its Tank-automotive and Armaments Command (TACOM), is engaged in an R&D program to develop a capability to produce replacement parts for equipment in the field at or near the point-of-need. This capability will enable the rapid repair and return to service of disabled equipment and address the military priority of weapon system readiness. The program, titled the Mobile Parts Hospital (MPH)*, is analogous to the Mobile Army Surgical Hospitals (M.A.S.H.). Since their inception, M.A.S.H. units have been effective at returning soldiers to health in the field. The MPH units are designed to do the same for Army equipment. The overall approach is to develop a mini-factory that can be deployed in the field to manufacture replacement parts in order to return disabled military equipment quickly to operational and combat-ready status. The Army faces a major challenge in equipment maintenance and combat readiness in today’s environment. It must continue to meet readiness standards while maintaining a large aging inventory of weapon systems. Increasing the effective service life of these legacy systems has led to an increased demand for spare parts, many of which are from product lines or manufacturers that no longer exist. Because the Army is purchasing fewer systems (and in significantly reduced volumes), suppliers assign lower priority to the Army’s production and schedule requirements, thereby increasing product lead times, particularly for forged and cast items. Additionally, the Army is poised to introduce several new systems into the inventory, many of which will pose logistical challenges when fielded. MPH is meant to address this challenge by significantly reducing the lead-time needed for replacement parts. The Army is actively developing advanced logistical systems to address these

challenges. The MPH will play an integral role in supporting these new systems, and will continue to support current supply systems. As the MPH program has progressed, three major developments have evolved 1) the mobile modules referred to as the rapid manufacturing system (RMS), 2) an agile manufacturing cell linked to the RMS via 3) the control and communications center. Technical Challenges The MPH manufacturing capability must meet the following requirements to be effective: be readily transportable, have a small logistics footprint, and be able to fabricate parts for which there is a critical need. These requirements translate into some important technical challenges. Transportable: The RMS is limited in size and weight and must be configured so that it is transportable by C-130 aircraft and/or the Army’s Palletized Load System. The number of pieces of equipment is limited by this volume requirement, but the system must maintain the capability to make a wide range of necessary parts. Consequently, the production equipment selected must be compact and sufficiently flexible to make a variety of parts. Moreover, ancillary operations such as heat treatment or plating need be kept to a minimum. Logistics Footprint: To maintain a small logistics footprint on the battlefield, the RMS must be self-sustaining in terms of power, environmental, communications and inventory needs. The current RMS prototype designs address these factors. Of particular significance is the need to minimize the amount and variety of material stock required to make the desired parts. This translates into the need for flexible fabrication processes to enable the production of a variety of parts from limited material stock that meet the functional engineering requirements of the part.

* www.mobilepartshospital.com The AMPTIAC Quarterly, Volume 6, Number 3

3

8’ Stowed Configuration

24’ Operational Configuration Fold Down Work Table

Fold Down Work Table LENS® Laser Chiller

LENS® Laser Chiller

LENS® Machine

HVAC Unit

Entrance Argon Container

LENS® Control

HVAC Unit

Entrance

Laser

Unit

Laser

Tool HVAC Cabinet

LENS® Machine

20’

Scanner

Scanner w/Engineering Workstation

Tool Cabinet

LENS® Control

HVAC Unit

Argon Container

Figure 1. Directed Material Deposition Module HVAC Tool Unit Cabinet Multitask Machining Unit (MAZAK®)

Air Compressor

8’ Stowed Configuration

Entrance

20’ HVAC Tool Unit Cabinet

Multitask Machining Unit (MAZAK®) 12’ Operational Configuration

Air Compressor Entrance

Figure 2. Multitask Machining Center Module

Critical Parts: Above all, a flexible fabrication capability is necessary to produce the range of critical parts required. During the development of the MPH, a parts selection process has been employed to identify select parts that: impact on systems readiness due to high cost, are not readily available (because of lack of suppliers), require a long lead time, and directly impact mission success. To provide MPH with the capability to make all these parts, a database is being built that houses the engineering and manufacturing data required to produce them. In addition, MPH integrates a satellite communication capability to transfer data from remote sites and includes a capability for reverse engineering when no data are available. The need for flexible manufacturing processes permeates the overall requirements for the MPH program. This flexibility has been facilitated by the adoption of advanced metal-cutting technology, as well as by the conversion of rapid prototyping technology into rapid manufacturing systems. Rapid Manufacturing System MPH mobile modules are designated as Rapid Manufacturing Systems (RMS). The RMS is a mobile manufacturing system that can produce parts rapidly near the point of need. The initial MPH utilized a 53-foot enclosed trailer, similar to a stan4

The AMPTIAC Quarterly, Volume 6, Number 3

dard road-going tractor-trailer arrangement. Transportation requirements however, dictated a smaller footprint. The current RMS prototype consists of two twenty-foot expandable ISO-standard (International Standards Organization) freight containers, each housing one primary piece of fabricating equipment. Each of these modules meets the size and weight requirements for a C-130 transport plane and can be moved overland on Army trucks. The first module contains a Directed Material Deposition (DMD™) machine that utilizes a patented process called Laser Engineered Net Shaping(™) (LENS®). LENS® can create a fully dense metal part from a computer aided design (CAD) model, which is converted to a standard triangulation language (STL) file. This module has a 20' by 8' footprint when stowed, but is designed to expand to 20' by 24' when deployed. (See Figure 1) After a part is built “near net shape” with this process, it is directed to the other machine – a 5-axis multi-task machining center for final finishing and dimensioning. This machine is housed in a 20' by 8' container that expands to 20' by 12' when deployed (See Figure 2). In addition to the major machines inside each module, there is also an engineering workstation with both engineering and manufacturing software, a satellite communications system, and a reverse engineering capability

LENS® is the registered trademark and service mark of Sandia National Laboratories and Sandia Corporation Laser Engineered Net Shaping(™) is a trademark of Sandia National Laboratories and Sandia Corporation

Figure 3. Directed Material Deposition Machine (Exterior View)

Figure 4. Directed Material Deposition Machine (Interior View)

including laser-scanning equipment to capture part geometry from old parts in need of replacement. When a request for a part comes to the RMS, its on-board databases are searched to determine if that specific part (or one similar to it) has been built before, and if the manufacturing data are available. To manufacture an article, the RMS must have a complete 3-Dimensional model of the requested part. If the required data were not available in the databases, on-board equipment could be used to create it – using the CAD/CAM software in combination with the non-contact laser scanner. Once a 3-D model is obtained, the file is converted to a format used by the LENS® manufacturing process. Laser Engineered Net Shape Process The LENS® process is considered a DMD process because powdered metal is directed into the path of a laser beam, where it is melted and then solidified on the workpiece, creating a component, particle-by-particle and layer-by-layer. The properties of metal parts created using this process can be made equivalent to and potentially better than those of wrought material of the same composition. In addition, the time to create a part using this process, compared to casting or forging, is greatly reduced (See Figures 3 & 4). The LENS® process is an emerging technology which is still maturing, thus both development and testing of LENS® continue in the MPH program. Controlled process optimization studies, employing design of experiments (DOE) methodology, are underway to relate process variables to resulting material microstructures and mechanical properties. In addition to relating the process variables to material properties, the experiments are designed to identify methodologies for increasing processing speed, improving surface finish, and maximizing the configurational capabilities of the process. Key process variables include laser power, powder feed rate, weld pool size, and beam path control strategy. A major asset of the LENS® process is the ability to make free form (without molds or dies) metal parts of high-density and

near-net-shape, enabling the MPH system to make a variety of parts. Inclusion of this process enhances MPH’s mobile processing capability. Because the process is based on powder technology, the material inventory can be dramatically reduced, as there is no need to store varieties of solid stock, in turn reducing the footprint of the facility. In addition, the ability of the process to consolidate a number of high performance alloy powders (e.g. high alloy steel, titanium) permits MPH to operate with a reduced material compositional inventory by building parts to better than original part properties. In this strategy, a single high performance alloy can be employed for a whole family of parts, yielding mechanical properties equal to or better than those in the original parts. Consequently, the compositional varieties of powder stock can be kept to a minimum in the materials inventory. Finally, the near-net-shape capability minimizes scrap, which consequently reduces the required total of the powder stock inventory. The ideal goal of the MPH program would be to carry a single powdered metal to the battle space that would meet all part property requirements. The Multitask Machining Center Post-processing of parts made by the LENS® process requires the use of a second piece of manufacturing equipment, the 5axis multitask milling machine to meet surface finish and tolerance specifications. The second ISO module of the RMS contains this 5-axis multi-task-machining center. This machine is a multi-axis mill that is set up primarily as a lathe. Work pieces such as gears and camshafts that normally would require separate turning centers, and both vertical and horizontal machining centers, can now be completely machined with efficiency and accuracy in the multi-task machining center. All axes are direct motor driven with no belts, pulleys, or gears, and tool exchange speed is nearly instantaneous (See Figure 5). The milling machine incorporates an integrated 64-bit computer numerical controller (CNC), color graphics display, and a simple programming language. In addition to having the capability of directly inputting machine code, the programmer The AMPTIAC Quarterly, Volume 6, Number 3

5

Figure 5. Multitask Machining Center

can also simply input the dimensions in a logical machining sequence (guided by the machine). The video display unit shows a shaded model of the work-piece for each stage, including a model of the cutting tools in action. This can be seen dynamically for an entire program prior to cutting material and while the actual machining is in progress. During programming, the controller immediately flags any x-y-z interference or an unfeasible operation as an error, and the corresponding program line is highlighted for correction. The 64-bit controller is capable of making tool path adjustments on the fly (OTF) to compensate for tool wear. Any manual adjustments (tweaking) done by the operator while the program is running can be recorded by the controller and immediately incorporated into the master program file, if desired. Another capability of this machine is the ability to record cutting path data in the event of tool breakage. These data (up to 5 motions) are recorded as the operator manually guides the tool away from the work-piece for changing. At the restart command, the stored cutting path data guides the tooling back to the interrupted stage position, and the original program continues. There is no need to return to the program beginning, saving significant time over conventional methods. Agile Manufacturing Cell Although the mobile RMS can successfully manufacture a variety of parts, it has repair part size and weight restrictions based on mobile process limitations. Several critical parts are either too large or have manufacturing requirements that are simply impractical for a mobile manufacturing environment. A prototype agile manufacturing cell (AMC) has been designed to support the mobile RMS. This flexible manufacturing would be located in the U.S. but would serve as a support system capable of handling these larger repair parts and processes not available in a mobile unit. The AMC will also be able to produce larger quantities of parts (50-100 each) in contrast to the RMS that is designed for low volumes on an as-needed basis. Control and Communications Center (C2) The C2 is the link for direct communication between the mobile systems and the U.S. based agile manufacturing cells. An essential function is the management and transmission of the engineering and manufacturing data required to produce replacement parts. 6

The AMPTIAC Quarterly, Volume 6, Number 3

The program’s engineering and manufacturing database serves as the manufacturing foundation for the MPH program. The part database is managed by product data management software (WindChill™) that is incorporated into the communications center. This software is designed to operate in a distributed engineering environment and consequently facilitates the control and transfer of information between multiple units. Producing parts on demand with a CAD/CAM data library provides a distinct advantage in timeliness. When the data are not available in the database, outside sources must be queried or reverse engineering is required. The MPH program continually identifies and adds parts to this database by gathering or creating manufacturing data. The ISO Standard for Exchange of Product (STEP) data format is the current, universally adaptable CAD language and manufacturing data format of choice, and is being utilized by the MPH program. The C2 utilizes a two-way satellite system to provide the communications link between the AMC, the RMS, and the Army’s established logistics systems. The system provides audio, video, and data exchange capabilities. The C2 satellite data transfer system can receive/send both CAD and/or CAM data from the RMS located anywhere in the world. The transferred data can be directly fed into either of the manufacturing machines or into the CAD/CAM workstation for further model definition and storage. The CAD/CAM database of parts and raw material procurement would logically be handled in a central location. TACOM-TARDEC (Warren, Michigan) is a likely location for the communications and control center, which will network the RMS fleet and agile manufacturing cells. Tentative plans are to merge the MPH C2 activities with those of the Emergency Operations Center already located at TACOM. Summary The development of a mobile manufacturing capability to produce replacement parts in the field will provide the Army with the ability to reduce spare parts inventory, reduce the cost and size of logistics operations, and provide for increased operational readiness. It is designed to complement existing supply systems and to integrate with the advanced logistic systems being developed for the Army of the future. Ultimately, the goal of MPH is to increase and enhance the effectiveness of our combat commanders. The MPH program is working with emerging technologies to achieve the manufacturing capabilities needed for the mobile environment. The most advanced rapid prototyping and manufacturing processes available today are not yet adequate to fulfill all the requirements of the MPH program, but great strides have been made in recent years. In the future, the MPH program can look forward to increased and broadened capabilities, producing better parts from a wider range of materials at faster build rates, with greater configurational range and accuracy. For more information on the MPH, visit www.mobileparts hospital.com.

Dr. Charles Sturrock, Dr. Edward Begley National Institute of Standards and Technology

Those of us who have dealt with the utilization, storage, or transfer of materials property data are well aware of the difficulties (and pitfalls) of handling such design-critical information. Whether in printed form (such as the MIL-Handbooks), or in electronic form (such as material property databases or material information systems), assimilating and displaying such data has proven to be a daunting task, and in some cases, an insurmountable one. The advent of the World Wide Web significantly advanced the state-of-the-art in information exchange. More than a decade since the initial revolution, a second generation of exchange languages has emerged, with greater flexibility and adaptability to the parochial terminologies of a disparate number of professions; such as chemistry, medicine, law, and even finance. For those of you who have wrestled with the challenges of material data standardization, data quality, or the handling of mass quantities of materials property information, the virtues and benefits of a materials-specific exchange language will be immediately obvious. For those not familiar with the topic, enjoy this first glimpse into the future! - Editor

MatML is an extensible markup language for the automated exchange of materials property data. In development since October 1999, MatML presently exists as a working draft, and is being revised in anticipation of its public release as a proposed recommendation. This article provides a brief description of MatML, including its history and sample markup. The vision of a future materials data exchange environment in which MatML will serve all parties with an interest in the promulgation of materials property data is also discussed. Introduction The World Wide Web is clearly one of the most remarkable developments of 20th century computing. From extremely modest beginnings in 1990 at CERN, the European Organization for Nuclear Research, the Web has grown exponentially to the point where today it is by far the largest source of information ever assembled. Like any information technology resource, the Web is supported by an extensive hardware and software infrastructure. One of the major components of the software infrastructure is hypertext markup language, or HTML, which is used to format nearly all of the content presently available on the Web. In recent years, the World Wide Web consortium (W3C – http://www.w3.org), the organization responsible for the various standards underlying the Web, including HTML, has found HTML to be insufficient for the processing and exchange of structured data via the Web. Responding to this

insufficiency, the W3C in February 1998 released the Extensible Markup Language (XML), which is actually a meta-language for defining markup languages used to describe domain-specific information. Various communities of specialists interested in exchanging data specific to their field have since come forward and developed markup languages based on XML. This article describes the efforts and results of the materials community in developing MatML, a markup language for describing and exchanging materials property data. What is MatML? In May 1999, a workshop entitled “Materials Data in the Internet Era” was held at the National Institute of Standards and Technology (NIST). The attendees recognized the tremendous potential of XML, which had been released the year before and was already beginning to spawn markup languages in various scientific and technical domains. In the fall of 1999, E.F. Begley of NIST assembled an international working group of about 40 materials data experts to determine the scope of MatML, a new language for materials property data, and formulate its vocabulary and syntax. Perhaps the best way to describe MatML is by example and in contrast to HTML. Table 1 contains an excerpt of data drawn from the NIST High-Temperature Superconducting Materials Database (WebHTS) as it would appear when rendered by a web browser. The AMPTIAC Quarterly, Volume 6, Number 3

7

Table 1. Tabular Display of Data Excerpt from WebHTS

Magnetic Field (Tesla) 0

Temperature (K) 3

Critical Current Density (kA/cm2) 3040

Table 2. MatML DTD Version 2.0 Complete Tagset

Associate Associations Bulk Details Characterization Chemical Composition Class Component Details Compound Concentration DataSource Data Type Dimension Details Dimensions

Elements Form Formula Geometry Graphs Material MatML–Doc Measurement Technique Name Notes Orientation Parameters Phase Composition

Processing Properties Property Details Qualifier Relationship Result Shape Source Specification Subclass Terms Units Value

The HTML markup for this table is as follows, with the tags shown in italics and the data shown in normal text: Magnetic Field (Tesla) Temperature (K) Critical Current Density (kA/cm2) 0 3 3040 Note that the tags in HTML specify only how the data are to be presented, and convey no descriptive content whatsoever. Here is the same excerpt in MatML markup: Critical Current Density kA/cm2 Journal Evaluated 3040 Magnetic Field 0 Tesla Temperature 3 K The descriptive nature of the MatML tags, such as , , and is plainly evident. Note also 8

The AMPTIAC Quarterly, Volume 6, Number 3

that the MatML version includes other information about the pedigree of the data, as conveyed by and . MatML, like all extensible markup languages, provides for any number of user-defined tags such as these. The descriptive nature and extensibility of the language renders it far more intelligible and malleable than non-descriptive fixed tagsets such as HTML. MatML presently exists as a working draft in XML Document Type Definition (DTD) format. Table 2 contains the complete tagset for the language. An annotated version of the entire DTD, along with many other MatML-related resources, can be found at the MatML website: http://matml.nist.gov/. June 2001 MatML Workshop Last summer, members of the materials data community convened at NIST at a workshop devoted to the technical and strategic future of MatML. The proceedings of this workshop can be found at the MatML website. In brief, the main findings of the workshop were as follows: • The development of MatML has been met with great interest, with frequent visits to the MatML website, downloads of the DTD, and inquiries to NIST and other parties that have contributed to the development of the language; • The extensibility of MatML makes it possible to address the full range of materials information needed by materials researchers, specialists, designers, and quality control analysts, as well as the full range of materials-related applications used by these professionals; • MatML markup is relatively easy for materials scientists and engineers to understand, as the code consists largely of materials and engineering terminology (see previous example), which has the key advantage of enabling materials specialists to review the markup in detail to ensure that all of the required parameters involving test methods, quality, reliability, and applicability of the data have been addressed; • A good set of illustrations demonstrating the power and flexibility of MatML should be established in order to advance the widespread adoption of MatML; • MatML must be tested thoroughly, involving individuals at various stages in the material development/application cycle ranging from materials specialists who select materials to designers responsible for addressing load-carrying capacity and failure limits. In addition, the needs of materials scientists exploring the fundamental structure and properties of materials, and of journal publishers disseminating materials information, must also be addressed; • At the appropriate time, MatML should be recognized officially and formally via registry with XML repositories such as xml.org and/or by standards organizations such as the Object Management Group, the American Society for Testing and Materials (ASTM), or the American National Standards Institute (ANSI);

• The overall effort should be linked with key industries, that might, via some of their applications, be in position to provide good tests for MatML and, assuming success, encourage or perhaps even require users of their information systems to use MatML as the data exchange medium. MatML - Present and Future MatML is presently being revised to be reissued in XML Schema representation. XML Schema provides robust data typing, including user-defined datatypes, and is much more suitable for application development than XML DTD representation. The next phase of MatML’s development will involve acceptance testing, and migrating to MatML Schema will provide sound support for such testing. With respect to acceptance testing, there are two principal kinds of XML applications: document publishing and document processing. Document publishing, which is the manipulation of information for human consumption, by necessity, comes first and will exhibit the strengths and weaknesses of MatML. In a sense, acceptance testing is already underway with the development of a compendium of markup examples, but wider participation is needed in order to assess MatML’s application to a broader and deeper range of materials. Future NIST testing might include markup of NIST WebSCD, NIST WebHTS, and NIST WebPDS, which are online compilations of structural ceramic, high temperature superconductor and property data summaries, respectively.[1] Document processing, which is the manipulation of information intended for software consumption, represents the second phase of MatML’s acceptance testing. While no specific plans have yet been identified, several projects are already exploring the use of MatML, including: • Green’s Function Research and Education Enhancement Network (Kent State University, Massachusetts Institute of Technology, NIST, (Web address: http://shreve.mcls.kent. edu/NSDLGreen/GREENProject.htm)); • femML – the finite element modeling markup language (J. Michopoulos, U.S. Naval Research Laboratory, (private communication)). Another important issue is education. It is not sufficient to publish MatML simply as a language specification. It is necessary to provide additional resources to promote understanding and adoption of the new language. This philosophy was behind the publication of MatML in an annotated format on the MatML website, where, in addition to the syntactic and semantic formalism of the language, English language explanations are provided as well as sample markup. Examples of other supporting resources for MatML found on the MatML website include links to sources of information about XML, and related expositions such as “XML, Element Types, DTDs and All That.” Finally, the compendium of markup examples will be extremely valuable for assisting users with learning how to apply MatML to their own data. Presently the three main strategic issues for MatML are: promotion and outreach, MatML registration, and long-term stewardship. Direct communications, journal publications, the

MatML website, conference presentations, and the MatML workshop proceedings can be used to reach data providers and consumers such as journal editors, handbook and database publishers, materials suppliers (data sheets), instrument manufacturers, and researchers. Registering MatML with the www.xml.org registry as well as the NIST registry/repository for XML may also provide broader exposure and access to MatML.[2] The MatML Steering Committee has been established to oversee and promote continued development and adoption of MatML, to coordinate acceptance testing activities, and to explore opportunities for MatML’s long-term stewardship. Its members include: J.G. Kaufman, Chairman, the Aluminum Assoc. (ret.) E.F. Begley, Lead Technical Expert, NIST F. Cverna, ASM International C. Grethlein, AMPTIAC S. McCormick, ESM Software D. Mies, MSC Software C. Nunez, CenTOR Software Corporation C. Sturrock, NIST M. Sullentrup, Boeing Company The MatML Steering Committee represents the interests of a much larger community of materials and computing professionals interested in MatML and its objective of automating the exchange of materials property data over the Web. This larger community is in turn kept abreast of developments via the MatML website and dedicated mailing list (for information contact [email protected]). At present, the materials data “marketplace” can be very chaotic and difficult for both providers and consumers. Until the arrival of MatML, there was no common exchange format, but instead hundreds of proprietary formats that resulted in wasteful duplication of effort and poor scale-up. Also plaguing the materials data community was (is) the absence of software interoperability, whereby one computer program, such as a finite element code, can automatically input material properties from another computer program or database without the necessity of human intervention. The present situation overall yields very inefficient data processing. MatML will serve the materials data marketplace as a common, public-domain materials data exchange format - a non-proprietary “Esperanto” for materials property data. With considerable flexibility and extensibility built-in as attributes of the language, MatML will provide for direct program-toprogram interoperability, efficient data processing, and rapid, reliable, and useful response to searches for materials data over the Web. Its widespread adoption will provide for powerful and malleable connection of the various materials data sources found on the Web today, as well as those to be added to the Web tomorrow. References [1] For more information, see the Ceramics WebBook at http://www.ceramics.nist.gov/webbook/webbook.htm [2] For more information, see the MatML website at http://matml.nist.gov The AMPTIAC Quarterly, Volume 6, Number 3

9

New AMPTIAC Products Available Now As the contents of the AMPTIAC Quarterly might suggest, we are in the business of promoting materials technology and technology transfer within the materials and defense communities. What you may not know is that AMPTIAC also offers a variety of products, printed and electronic, to promote technical awareness and serve as an educational resource for the community. Our complete and interactive product catalog, along with descriptions of our other technical services, are available for viewing at the AMPTIAC website, http://amptiac.iitri.org/ ProductsAndServices. If you have any questions about any of our products, or if you would like to order one or more of them, please contact our Product Sales Manager, Gina Nash at (315) 339-7047 or e-mail at [email protected]. This past year has been an exceptional one for AMPTIAC, with the publication of a wide range of new technical products. Regardless of your specific discipline, there is an AMPTIAC product for you. Among our most notable publications this year:

Other Recent Products

New Releases

Computational Materials Science (CMS) – A Critical Review and Technology Assessment AMPTIAC surveyed DOD, government, and academic efforts currently studying materials science by computational methods and from this research compiled this report. It provides and indepth examination of CMS and describes many of the programs, techniques, and methodologies being used and developed. The report was sponsored by Dr. Lewis Sloter, Staff Specialist, Materials and Structures, in the Office of the Deputy Undersecretary of Defense for Science and Technology. BONUS MATERIAL: Dr. Sloter also hosted a workshop (organized by AMPTIAC) in April 2001 for the nation's leaders in CMS to discuss their current programs and predict the future of CMS. The workshop proceedings comprise all original submitted materials for the workshop - presentations, papers, minutes, and roundtable discussion highlights and are included with purchase of the above report.

New! Textile Preforms for Composite Material Technology Newly released, this publication is the first and only one of its kind – A panoramic and thorough examination of fiber/textile preform technology and its critical role in the development and manufacture of high-performance composite materials. This product was prepared in collaboration with Drexel University and authored by Dr. Frank Ko, the Director of Drexel’s Fibrous Materials Research Center. Dr. Ko is one the world’s foremost authorities on fibrous preforms and textile technology. Order Code: AMPT-19

Price: $100 US

New! Material Selection and Manufacturing for Spacecraft and Launch Vehicles A first of its kind publication, this State-of-the-Art Review provides a comprehensive overview of the unique requirements, problems, and opportunities faced by engineers designing and manufacturing spacecraft and launch vehicles. The book is authored by Dr. Carl Zweben, a major leader in the materials and space communities over the past several decades. While all material aspects of spacecraft and launch vehicles are addressed in this work, a special emphasis is placed on the unique qualities of materials used in space and highlights differences between them and their counterparts used in air/land/sea applications. Order Code: AMPT-22

Price: $100 US

New! YBa2Cu3Ox Superconductors – A Critical Review and Technology Assessment An up-to-date report highlighting recent research on the processing of YBa2Cu3O(7-δ) (YBCO) superconducting materials to produce high critical current densities. The processing of powders through the fabrication of bulk materials and the deposition of films is covered, along with advantages and disadvantages of the various manufacturing methods. Problems in finding suitable substrate materials along with the associated property characterizations are presented. Some applications for YBCO superconductors under research, and ones that have been realized, complete the report. Order Code: AMPT-27 10

The AMPTIAC Quarterly, Volume 6, Number 3

Price: $50 US

Applications of Structural Materials for Protection from Explosions This State-of-the-Art Review provides an examination of existing technologies for protecting structures from explosions. The report does not discuss materials and properties on an absolute scale; rather, it addresses the functionality of structural materials in the protection against blast. Each chapter incorporates information according to its relevance to blast mitigation. For example, the section on military structures describes concrete in arches, and concrete in roof beams for hardened shelters. The discussion on concrete is not limited to materials only; rather, it addresses the issue of structural components that incorporate concrete, and describes the materials that work in concert with the concrete to produce a blast-resistant structure. The report also illustrates various materials used for concrete reinforcement. Order Code: AMPT-21

Order Code: AMPT-25

Price: $100 US

Price: $65 US

Blast and Penetration Resistant Materials This State-of-the-Art Review compiles the recent and legacy DOD unclassified data on blast and penetration resistant materials (BPRM) and how they are used in structures and armor. Special attention was paid to novel combinations of materials and new, unique uses for traditional materials. This report was sponsored by Dr. Lewis Sloter, Staff Specialist, Materials and Structures, in the Office of the Deputy Undersecretary of Defense for Science & Technology. BONUS MATERIAL: Dr. Sloter also hosted a workshop in April, 2001 (organized by AMPTIAC) for selected experts in the field of BPRM and its application. The workshop focused on novel approaches to structural protection from both blast effects and penetration phenomena. Some areas covered are: building protection from bomb blast and fragments, vehicle protection, storage of munitions and containment of accidental detonations, and executive protection. The proceedings of this workshop are included with purchase of the above. Order Code: AMPT-26

Price: $115 US

Material E

A

19 15

A MPTIAC

S

E

Jeffrey Guthrie, Brigitte Battat and Chris Grethlein AMPTIAC, Rome, NY

ACCELERATED CORROSION TESTING

“Corrosion is Corrosion” – we have heard that all too often at the AMPTIAC inquiry desk. In our experience, corrosion is one of the most misunderstood and mischaracterized forms of material degradation. Consequently, corrosion analysis and mitigation methods tend to be some of the most misapplied. This MaterialEASE is the third and final installment excerpting a new AMPTIAC State-of-the-Art Review entitled “Accelerated Testing: A Methodology for Determining Life and Degradation of Materials”. It will be published in the near future. This article examines the fundamentals of accelerated corrosion testing, the associated pitfalls therein, and what tools today’s materials engineer will have to battle corrosion.

Background The most accepted definition for corrosion is the destruction of material due to a chemical reaction of the material with its environment. Generally, this destruction takes place on its surface in the form of material dissolution or redeposition in some other form. Metallic systems are the predominant materials of construction, and as a class, are generally susceptible to corrosion. Consequently, the bulk of corrosion science focuses upon metals and alloys. Corrosion does occur in polymers and ceramics but the mechanisms are quite different from those of metals. As such, this article limits itself to a discussion of metallic corrosion.

The Problem: As recently as 1995, a major study concluded that the cost impact of corrosion to the U.S. economy totaled nearly $300 billion annually.[1] This represents nearly 4% of the Gross Domestic Product, and is a conservative estimate at best – indirect costs, which are substantial and mostly unreported, would greatly inflate the total estimated cost. Estimates suggest that about two-thirds of present corrosion-related costs are unavoidable. In addition, corrosion affects our Nation’s force effectiveness and readiness levels through the diminished safety and reliability of structures, mechanisms and electronics. In many cases, corrosion is the life-limiting factor of a component. Corrosive failures can occur unexpectedly at the worst possible moment. Corrosion testing can consume enormous blocks of time; particularly in the case of outdoor atmospheric tests. Unfortunately, the timescales involved in such tests preclude the opportunity for proper materials selection. In typical circumstances, new systems may be halfway through their lifecycle before real data on the fielded system would indicate any corrosion problems. Under the right conditions, accelerated testing may yield data beneficial in selecting the most corrosion resistant materials for an application. Accelerated testing isn’t limited to the design stage of a system’s lifecycle, but can be used to provide in-field support, as the emergence of sudden corrosion problems on fielded systems requires quick answers. Managing corrosion in structural components and critical systems, to extend service life and ensure reliability, is paramount. Effective corrosion control requires meaningful test data in a reasonable time frame such that it may be employed to influence materials selection and protection efforts. There are presently hundreds of existing accelerated corrosion test methods and no doubt there will be many more in the future.

The Electrochemical Process Corrosion consists of an oxidation reaction and a reduction reaction at the surface of the corroding material. The oxidation reaction generates metal ions and electrons; the electrons are then consumed in the reduction reaction. For environments with water present including moisture in the air, the electrons are consumed by converting oxygen and water to hydroxide ions. In iron and many iron alloys, these hydroxide ions in-turn combine with iron ions to form a hydrated oxide (Fe(OH)2). Subsequent reactions form a mix of magnetite (Fe3O4) and hematite (Fe2O3). This red-brown mixture of iron oxides is rust. The figure below illustrates the basic oxidation/reduction reaction behind corrosion. The higher the ionic conductivity, the quicker this reaction takes place. This is why water containing electrolytes, such as salt, is far more damaging. Another key point to note is that reducing the amount of dissolved oxygen in solution directly can inhibit corrosion. However, many other reduction reactions can consume the electrons.

OH

Fe+2

Ir on

OH

OH

OH

H 2O

O2

e

e

H2O e

Fe+2 e

Figure 1. Electrochemical Process

The AMPTIAC Quarterly, Volume 6, Number 3

11

Material E

A

S

E

Table 1. Traditional Corrosion Classifications[3]. Uniform (General) .. Occurs evenly over the entire surface. The rate of corrosion is often presented as a weight loss. Uniform corrosion is very predictable, and is the basis of most corrosion prediction equations. Galvanic .................. Corrosion caused by an electrochemical reaction between two dissimilar metals when placed in contact or electrically connected with each other in an electrolyte. Crevice .................... A localized form of corrosion which occurs within the stagnant zones created by the interfaces between two surfaces. Crevice corrosion can occur under washers, gaskets, debonded coatings, surface deposits, sealing rings, rivets, sleeves, surface scale, switch contacts, clamps, etc. The diverse environmental factors causing crevice corrosion make it almost impossible to simulate field conditions in the laboratory. Predicted life from accelerated tests is also unlikely. Pitting ...................... Pitting corrosion is a localized form of corrosion that proceeds with minimal overall metal loss, making it difficult to detect. Consequently, it is difficult to predict, and its rate of attack is hard to quantify, as its reaction rate is not quantifiable. Intergranular .......... This phenomenon occurs preferentially at grain boundaries, usually with slight or negligible attack on the adjacent grains. Selective Leaching .. One element is preferentially removed from an alloy, leaving a residue (often porous) of the elements that are more resistant to the particular environment. It is also called de-alloying or parting. Common forms of selective leaching are decarburization, decobaltification, denickelification, dezincification, and graphitic corrosion. Erosion .................... An action involving both corrosion and erosion in the presence of a moving corrosive fluid, leading to the accelerated loss of material. Stress ...................... Induced by the joint influence of mechanical stress and corrosion processes. Static-tensile stress causes stress corrosion cracking (SCC). The stresses initiating SCC damage are in the form of applied mechanical stress or residual stress. Manufacturing processes such as welding, bending, cold deformation (cold working), or electroplating of metallic components can bring about residual stresses. Differences in the coefficients of thermal expansion (CTE) of the materials in contact increase the risk of SCC. Corrosion fatigue is often considered to be a form of SCC. Cyclic loading brings about corrosion fatigue.

Table 2. Forms of Corrosion – ASM Classifications[4]. General Corrosion: Corrosive attack dominated by uniform thinning • Atmospheric Corrosion • Galvanic Corrosion • Stray-Current Corrosion • General Biological Corrosion • Molten Salt Corrosion • Corrosion in Liquid Metals • High-Temperature Corrosion

12

Localized Corrosion: High rates of metal penetration at specific sites • Crevice Corrosion • Filiform Corrosion • Pitting Corrosion • Localized Biological Corrosion

The AMPTIAC Quarterly, Volume 6, Number 3

Metallurgically Influenced Corrosion: Affected by alloy chemistry & heat treatment • Intergranular Corrosion • Dealloying Corrosion

Mechanically Assisted Degradation: Corrosion with a mechanical component • Erosion Corrosion • Fretting Corrosion • Cavitation and Water Drop Impingement • Corrosion Fatigue

Environmentally Induced Cracking: Cracking produced by corrosion, in the presence of stress. • Stress-Corrosion Cracking (SCC) • Hydrogen Damage • Liquid Metal Embrittlement • Solid Metal Induced Embrittlement

A D VA N C E D M AT E R I A L S

AND

PROCESSES TECHNOLOGY

A MPTIAC

Table 3. ASTM Standard Test Methods for Various Environments Environment causing corrosion Types of Tests Fresh Water 39 ASTM Standards Seawater Environments

ASTM B117, Salt Spray, Fog Test (Oldest and Most Widely used Cabinet Test)

General Environments (e.g., atmospheric, galvanic)

ASTM G31 (Immersion Corrosion Testing), ASTM G50 (Atmospheric Corrosion Testing)

The corrosion of iron illustrates that many reactions can occur. Similar to iron, aluminum forms an oxide scale; however, the oxide scale formed (alumina) is cohesive, adherent and inhibits further corrosion. The oxide film formed ennobles the metal and reduces electrical current density by as much as six orders of magnitude. Materials like aluminum exhibit three corrosion regions: a. Active Region -Increases in the oxidation potential lead to increasing corrosion rates. This region corresponds to the beginning of corrosion. b. Passive Region - Increasing the oxidation potential past the active region reduces the corrosion rate. c. Transpassive Region - Increasing the oxidation potential past the passive region, increases corrosion rate. If the oxidizing potential is raised to high enough levels the oxide film breaks down and the material begins to actively corrode again. Given the many reactions that can occur and the corrosion regions possible, accelerating the corrosion process is difficult without altering the corrosion mechanism. Categorizing corrosion forms into specific groups is not straightforward, because such groups have overlapping characteristics, which influence the initiation and propagation of corrosion. However, corrosion is often categorized by how it manifests itself, as described in Table 1, Traditional Corrosion Classifications. Classes of Corrosion ASM offers a complementary classification of corrosion that can be useful for designers. This classification, shown in Table 2, categorizes various corrosion processes by mechanism of attack. Inspection of this table indicates that some forms of corrosion affect (relatively) large areas as indicated in the section addressing “General Corrosion” while other mechanisms are very local in nature. Other corrosion mechanisms are facilitated by metallurgical factors while still others rely upon the movement of two structures relative to each other or a structure or component subjected to moving fluids or slurries. The final category identifies corrosion mechanisms that result from the inspection of a “stressed” material in the presence of a corrosive agent. Some of these mechanisms can be easily dealt with or mitigated through proper design and materials selection decisions. Accelerated testing can be a useful tool in this process as long as the investigator realizes the intent of the test. Accelerated tests can seldom be used to accurately predict the life of a structure. Rather, they more appropriately give insight in a qualitative fashion to how materials will behave when subjected to an environment. In either classification system, the corrosion categories overlap and often influence each other. Consequently, the classification of corrosion into groups may be gradually losing its significance as understanding of basic corrosion mechanisms continues to grow. Because of its intricacy, accelerated testing for corrosion is difficult to implement, and most types of corrosion do not lend themselves to prediction of performance through accelerated tests.

Corrosion Testing Typically, all conventional corrosion testing requires long testing times and is expensive. There are three major types of corrosion tests: Laboratory, Field, and Service Testing. Naturally, service testing provides the highest fidelity results followed by field testing. However, the service and field tests have limited prospects for accelerated testing. Accelerated corrosion testing, like any other form of accelerated testing, becomes increasingly more realistic as the laboratory environment approaches that of the service environment. To account for the variation of corrosion properties with configuration, several other standardized tests are performed for particular corrosion types, such as uniform attack, pitting, stress corrosion, and crevice corrosion. There are many ASTM and ISO publications describing standard test methods for different types of corrosion testing in various atmospheres. Table 3 shows several common tests addressing a variety of environments. Before testing Before implementing any corrosion test, care must be taken when selecting the corrosive media and preparing the test specimen. Accelerating tests often represent the worst-case scenario, requiring a very aggressive corroding factor. The onset of corrosion may seriously accelerate damage in an entirely unpredictable manner. Caution must be exercised to make certain the corrosion mechanism is not altered. Consequently, most accelerated tests should not be used to predict life or corrosion rates. These tests are typically qualitative, and the information obtained from them is best used to down-select the most appropriate materials for use in specific applications. Selection of Corrosive Media: The corrosion process largely depends on the corrosive environment. A thorough study of this environment is suggested before planning any corrosion test. In addition to analyzing the characteristics of that environment, make sure that the test does not introduce additional corrosion mechanisms. For instance, inadvertently using a bolt and test specimen which create a galvanic couple will lead to spurious results. If the corrosion experiment has an extremely stagnant electrolyte for a crevice corrosion test, the results may not show as aggressive an attack as desired since the amount of dissolved oxygen in a liquid environment greatly impacts the corrosion rate. Test Specimen Preparation: In addition to carefully selecting the corrosive environment, the corrosion specimen must be prepared in a fashion indicative of the service material. Corrosion behavior is influenced by the slightest variation in metallurgical structure and surface preparation - thus care must be taken while preparing the surface of the testing specimens. Avoid excessive heating of the surface from grinding operations to remove scale and remove excessive cold worked surfaces from sample machining, like those found near sheared surfaces or stamped edges. The AMPTIAC Quarterly, Volume 6, Number 3

13

Material E

A

S

E

Table 4. Summary of Specific Laboratory Corrosion Tests and Their Corresponding Standards Electrochemical Methods ASTM Standard Electrochemical polarization resistance G-59, D-2776 Electrochemical impedance G-106 Cyclic potentiodynamic polarization G-61 Galvanostatic polarization D-6208 Scratch-repassivation D-6208 Immersion Tests Total Immersion G-31 Partial immersion G-31 Intermittent immersion G-44 Salt Spray (Fog) Testing Neutral B-117 Acetic Acid G-85 Annex A1 Cyclic Acidified G-85 Annex A2 Cyclic Seawater Acidified G-85 Annex A3 Cyclic SO2 G-85 Annex A4 Dilute Electrolyte Cyclic Fog Dry G-85 Annex A5 Copper-Accelerated Acetic Acid B-368

0.8 Extrapolation of Initical Corrosion Rate

0.7

Average Penetration, mm

0.6

0.5

0.4

2-Point Estimate of Linear Rate

0.3

Actual Corrosion Curve

0.2

0.1

0 1

2

4

8 Exposure Time, years

16

Figure 2. Example of How Short-Term Corrosion Test Results Can Be Misleading.[7]

14

The AMPTIAC Quarterly, Volume 6, Number 3

Acceleration Testing does not have to correlate exactly to the service environment as long as the corrosion mechanism remains the same. However, the results of accelerated testing should correlate to results from more reliable sources, e.g., service experience and field testing. The method of accelerating corrosion testing will depend upon the material being examined, the environment and the type of corrosion mechanism. Several common methods for acceleration include increasing: • Aeration for immersion tests • Temperature (The most common and greatest accelerating factor for many corrosion mechanisms.) • Pressure for stress corrosion cracking tests • Acidity in immersion tests • The amount of NO2 and SO2 for atmospheric tests • Relative humidity for atmospheric tests • Impingement velocity in erosion corrosion tests • Current in electronic parts prone to corrosion Additionally for aqueous corrosion, the corroding electrolyte and the ionic conductivity help determine the rate of corrosion attack. Even when accelerated, corrosion tests can be inherently slow; hence, testers must exercise control of these variables. Besides altering the main corrosion mechanisms, gross variations of all these factors can alter the kinetics of several intermediate corrosion reactions leading to spurious results. Since most laboratory corrosion tests are accelerated tests, careful attention to, and meticulous execution of, their procedure is necessary to obtain useful results. Standard testing procedures have been set up for the various laboratory tests. These are described in some detail in Reference 5, and an outline of some of the specific tests and their corresponding ASTM standards is given in Table 4. There are many other factors affecting corrosion rates, which lead to countless test methods. There are fifteen ASTM standards on controlled humidity tests alone. Moreover, each test includes a range of test conditions. ASTM B 117 allows the concentration of NaCl solution to range from 3.5% to 20%. The goal is to accelerate the test as much as possible while keeping the test corrosion mechanism the same as the service corrosion mechanism. Interpreting the results In order to correctly interpret the results of any corrosion test, a complete pedigree of material information and testing parameters should accompany the test results. Variations in the material’s chemical composition, fabrication history, shape and size of the specimen all affect the corrosion rate. For example, the corrosion rate of wrought versus cast alloy will likely be different. Classifying materials in broad categories such as “Stainless Steel” or “Aluminum” is a mistake. The changes in specimen weight versus time are most often used to calculate corrosion rates. They are the easiest measurement to make. Additionally, electrochemical polarization methods for measuring corrosion rate are amenable to aqueous corrosion.[4] Other methods, such as directly measuring a “depth-of-penetration,” can be difficult. Unfortunately, measuring weight loss is unsuitable for localized corrosion effects such as pitting, crevice and intergranular attack since total weight loss can be minimal while local damage can be severe.

A MPTIAC A D VA N C E D M AT E R I A L S

For general corrosion, (uniform thinning) the corrosion rate is proportional to the weight loss and inversely proportional to the area, exposure time and density of the specimen. (Use the original area, not the reduced area, to calculate rates.) Even with general corrosion, care must be taken while testing to correlate a weight loss to a corrosion rate. For example, many materials produce a scale. If this scale is adherent but does not protect the base metal, it must be removed before weighing. Adherent scales may exhibit a weight growth even though the material is severely damaged. Note, the sample should be carefully examined before cleaning to gain insight into the corrosion mechanism. For example, adherent deposits or encrustations may cause pitting or crevice corrosion. To compound the damage assessment, the corrosion rate often changes with time due to film formation on the material’s surface. This change in corrosion rate can even undergo cyclical acceleration and deceleration if a scale thickens and sloughs off. For example, copper immersed in tropical seawater corrodes at a rate that is proportional to the square root of exposure time and not linear to it (Figure 2). Curves α and β were calculated from short-term test data (6-months and 2-year rate respectively). Curve γ is the actual corrosion rate, which resembles a long-term corrosion rate at short exposure times. On the other hand, low-carbon steel immersed in a similar environment initially forms a protective film after which the corrosion rate becomes almost linear with time. The ASM Handbook on Corrosion[5,6] recommends that at least three replicate sets of specimens should be tested through exposure at increased duration, to validate the data. Pitfalls of Accelerated Testing With any accelerated testing, caution must be exercised to gain meaningful results. This is particularly true with accelerated corrosion testing, where the “pitfalls” are numerous. The following describes some of the more prominent ones. Multiple Failure Modes complicate the design of accelerated testing. New failure modes may appear that can cause the component or the material to degrade or fail. The example mentioned earlier describing scale creating sites for pitting or crevice corrosion is a common case of multiple mechanisms at work. Test specimens which fail by an alternate mechanism not experienced in the field must be considered invalid. Misapplied Models are responsible for many of the errors of corrosion rate and life prediction. Many of the corrosion rate formulas are frequently misused. Most common among these is the application of predictive equations (often derived from uniform corrosion phenomena) to characterize and predict the behavior of localized phenomena, such as pitting or crevice corrosion. While some debate in this area still exists, one thing is clear: uniform corrosion models cannot and should not be applied to localized phenomena. Model Uncertainty is tremendously amplified via extrapolation of results from the accelerated test. For example, small sample size, sample shape, small numbers of failures, and large degrees of extrapolation from available data all combine to create great uncertainty. Unquantifiable Corrosion factors are major contributors to the uncertainty of models and their frequent misapplication. Certain factors render components much more susceptible to corrosion, such as geometry (crevice corrosion, pitting) or welds (selective leaching, SCC). Such effects, while substantial, are not readily quantifiable. Consequently, data from accelerated

AND

PROCESSES TECHNOLOGY

testing cannot provide reliable life predictions in such cases, and is instead limited to qualitative assessments. Multiple Factors Change Degradation when Multiple Time Scale is Involved. Most failures are the result of several factors. If only one of these factors is accelerated, it may invalidate the contributors of non-accelerated factors. A classic “snake in the grass” is the case of immersing a steel specimen in water near the boiling point. As the temperature approaches the boiling point the corrosion rate decreases since the free oxygen in the solution is driven off. While accelerating one failure mode, a different failure mode (that is usually the first to appear in the field) is masked. Acceleration of Specific Variables May Actually Decelerate Failure. For example, in the case of accelerated testing of a circuit-pack, the test increased temperature, thus lowering humidity and inhibiting corrosion. The result was fewer test failures as compared to field conditions. Prototype Articles for Laboratory Testing May Vary from Production Units. Materials and parts used to build prototypes may vary from the ones used in production, partly due to different levels of cleanliness, as well as the extra attention paid when building them. Test units should be taken from the field, and should use raw materials from actual production. Manufacturing conditions should be closely simulated in the accelerated test. These pitfalls illustrate that, among other difficulties, the field environment is not as carefully controlled as the laboratory. Because of this, it is difficult to design an accelerated test that can predict field reliability. Climate, which is supremely stochastic and essentially random, greatly affects the environmental conditions, thus rendering test variables highly unpredictable. Some have attempted to “average” out environmental conditions for accelerated life tests in the hope of reproducing the variable effects of climate. However, experience has shown that “smearing” environmental effects in accelerated tests does not provide good prediction of field reliability. Conclusion Corrosion is a complex electrochemical process dependent upon a myriad of factors. Be extremely cautious when devising any accelerated test technique or using a corrosion rate as gospel for predicting lifetimes. Accelerated corrosion testing is a necessary and powerful tool when used to: (a) help with materials selection as a relative indicator of corrosion resistance, (b) help examine potential environments for new materials, and (c) help determine corrosion control strategies of fielded items. References [1] Kuruvilla, A.K., Life Prediction and Performance Assurance of Structural Materials in Corrosive Environments - A State of the Art Report (AMPT-15), AMPTIAC, 1999. [2] Fontana, M.G., Corrosion Engineering, 3rd Edition, McGraw-Hill Book Company, 1986, pp 39-152. [3] Metals Handbook Ninth Edition, Vol. 13, ASM International, 1987, pp. 79-189 [4] Scully, J.R., Taylor, D.W., “Electrochemical Methods of Corrosion Testing,” Metals Handbook, Vol. 13, ASM, 1987, pp. 212-228 [5] Metals Handbook Ninth Edition, Vol. 13, ASM International, 1987, pp. 191-317 [6] Baboian, Robert, Editor, “Corrosion Tests and Standards: Application and Interpretation,” ASTM, 1995. The AMPTIAC Quarterly, Volume 6, Number 3

15

Mark Your Calendar Workshop on Synthetic Aperture Radar Tech 10/16/02 - 10/17/02 Redstone Arensal, AL Contact: Sherry Starling University of Alabama Huntsville, AL 35899 Phone: (256) 876-2628 Email: [email protected]. army.mil Web Link: smaplab.ri.uah.edu/SAR02

Workshop on Lean Manufacturing 11/20/02 - 11/21/02 Redstone Arsenal, AL Contact: Sherry Starling University of Alabama in Huntsville Huntsville, AL 35899 Phone: (256) 876-2628 Email: [email protected]. army.mil Web Link: smaplab.ri.uah.edu/LM02/

Refractory Ceramics Division Fall Mtg

DMC-2002 Defense Manufacturing Conference

10/16/02 - 10/18/02 Pittsburgh, PA The American Ceramic Society PO Box 6136 Westerville, OH 43086-6136 Phone: (614) 890-4700 Fax: (614) 899-6109 Email: [email protected] Web Link: www.ceramics.org

12/02/02 - 12/05/02 Dallas, TX Universal Technologies Corp. 1270 North Fairfield Road Dayton, OH 45432 Phone: (937) 426-2808 Fax: (937) 426-8755 Email: [email protected] Web Link: http://www.dmc.utdayton.com/

INTEGRAM - Designing Manufacturable MEMS for Product Integration

Workshop on Nano and Microsystems Technology and Metrology

10/29/02 Ulm, Germany 10/30/02 Grenoble, France 11/01/02 Malver, United Kindom Contact: Gail Messari Coventor, Inc. 4001 Weston Parkway Cary, NC 27513 Phone: (919) 854-7500 x102 Fax: (919) 854-7501 Email: [email protected] Web Link: www.coventor.com

12/04/02 - 12/05/02 Redstone Arsenal, AL Contact: Sherry Starling University of Alabama in Huntsville Huntsville, AL 35899 Phone: (256) 876-2628 Email: [email protected]. army.mil Web Link: smaplab.ri.uah.edu/NMTM02/

ShipTech 2003 - A Shipbuilding Technologies Information Exchange 01/16/03 - 01/17/03 Biloxi, MS Contact: Tricia Wright, Event Coordinator Phone: (814) 269-2567 Email: [email protected] Web Link: www.ncemt.ctc.com

27th Annual Cocoa Beach Conference & Expo. on Advanced Ceramics & Composites 01/26/03 - 01/31/03 Cocoa Beach, FL The American Ceramic Society PO Box 6136 Westerville, OH 43086-6136 Phone: (614) 890-4700 Fax: (614) 899-6109 Email: [email protected] Web Link: www.ceramics.org

High Temple Workshop XXIII

1st Internatl Defence Nanotechnology Conf 10/31/02 - 11/01/02 London, United Kingdom Contact: Alexander Giles Defence Event Management Ltd. Southbank House, Black Prince Road London SE1 7SJ U. Kingdom Phone: +44 (0) 208 297 8005 Fax: +44 (0) 709 216 2703 Email: [email protected] Web Link: www.dem-ltd.com

Workshop on Life Cycle Systems Engineering 11/06/02 - 11/07/02 Redstone Arensal, AL Contact: Sherry Starling University of Alabama Huntsville, AL 35899 Email: [email protected]. army.mil Web Link: smaplab.ri.uah.edu/lcse02

16

The AMPTIAC Quarterly, Volume 6, Number 3

5th International Military Sensing Symposia

02/10/03 - 02/13/03 Jacksonville, FL Contact: Jim Sutter NASA Glenn Research Center M/S 49-3, 21000 Brookpark Rd Cleveland, OH 44135 Phone: (216) 433-3226 Fax: (505) 846-8265 Email: [email protected] Web Link: www.namis.iitri.org

7th International Conference Commercialization of Military and Space Electronics

12/09/02 - 12/12/02 Gaithersburg, MD Contact: International MSS Committee c/o Infrared Information Analysis (IRIA) Center Veridian Systems Division, Inc PO Box 134008 Ann Arbor, MI 48113-4008 Phone: (734) 994-1200 Fax: (734) 994-5550 Email: [email protected] Web Link: www.iriacenter.org

02/10/03 - 02/13/03 Los Angeles, CA Contact: Dale Stamps or Leon Hamiter; Composites Technology, Inc. Huntsville, AL 35801 Phone: (256) 536-1304 Fax: (256) 539-8477 Email: [email protected] [email protected] Web Link: www.cti-us.com

ASIP 2002 (USAF Aircraft Structural Integrity Program Conference)

03/02/03 - 03/06/03 San Diego, CA Contact: TMS Customer Service; TMS 420 Commonwealth Dr. Warrendale, PA 15086 Phone: (412) 776-9000 x 253 Fax: (412) 776-3770 Email: [email protected] Web Link: cms.tms.org

12/10/02 - 12/12/02 Savannah, GA Contact: Jill Jennewine Universal Technology Corporation 1270 North Fairfield Road Dayton, OH 45432-2600 Phone: (937) 426-2808 Fax: (937) 426-8755 Email: [email protected]; Web Link: www.asipcon.com/

132nd TMS Annual Meeting & Exhibition

W.C.L. Shih, G.L. Fitzpatrick, PRI R&D, 25500 Hawthorne Blvd. #2300 Torrance, CA 90505

Introduction Multiple pressures are driving both commercial and military airframes to stay in service much longer than their designed service life. Accordingly, inspection requirements to insure aircraft airworthiness have created a need for cost effective NDI techniques that are accurate, reliable and easy to use.

Figure 1. Using the MOI*

spurred companies like PRI Research & Development (PRI) to bring their budding technologies to the marketplace. One such technology is the Magneto Optic Imager (MOI), introduced by PRI to the aircraft maintenance world in 1990 at the Farnborough Air Show in the United Kingdom. Application of the instrument is shown in Figure 1. It consists of a control unit, a handheld imager and a head-mounted video display. Since its introduction, use of the MOI has steadily increased particularly in areas where it clearly outperformed traditional methods such as eddy-current spot probes and sliding probes. Multiple tests have shown the MOI to be a fast and reliable method of finding surface and subsurface cracks and sometimes corrosion in metal aircraft skins. Since its introduction, many changes and improvements to the instrument have been made. These changes have been driven by the changing inspection priorities and have resulted in the new MOI 308 currently used today (Figure 2). The number of procedures by major aircraft manufacturers continues to increase, allowing maintenance personnel to use the instrument in many more inspections, increasing the cost effectiveness of the instrument and their acceptance of its place in the NDT arsenal.

Studies made it clear that aircraft kept on the ground during long maintenance cycles dramatically increased the expense of air transport. Maintenance groups not only needed new and better equipment, they also needed a process to implement these new technologies into the maintenance mainstream. Magneto-optic imaging is one such technique, and has gained wide acceptance for detection of both surface and subsurface defects. Similar requirements apply to military aircraft as well if not more so. Many military airframes are expected to fly up to, and in some cases more than, twice their design life. Some flight crews may be operating the same airframes that flew 308/7 Imager their fathers or even grandfathers! Inspections for these aging aircraft must be capable of detecting surface and subsurface defects, including cracks and corrosion. This article discusses the development of the magneto-optic technology and its current inspection applications in both commercial and military aircraft maintenance programs. An incident involving an Aloha Airlines Boeing 737, where a large section of the plane’s Low Frequency skin peeled off during flight, defined the need to Eddy-current Attachment perform careful periodic inspections. It also

Personal Video System

308/3 Imager

Control Unit

Figure 2. The MOI 308 System * All images in this article courtesy PRI Research & Development The AMPTIAC Quarterly, Volume 6, Number 3

17

Light Source Analyzer Polarizer

The Technology: Its Growth and Change The MOI uses a combination of an innovative eddy current induction method to induce magnetic fields in defects and magneto-optics to form images of the magnetic fields associated with the defects. These real time field images closely resemble the defects themselves. The MOI is able to image through paint and other surface coverings in real time and displays results as visual images on a heads-up display and/or an ordinary TV monitor. The instrument is hand-held, portable, requires minimal training, and greatly increases the speed and reliability of inspection. Results may be videotaped, printed using a video printer or captured digitally. The magneto-optic/eddy current nondestructive testing instrument is based in part on the principles of Faraday magneto-optic rotation. In 1845 British physicist Michael Faraday first observed the effect when linearly polarized light was transmitted through a piece of glass placed in an external magnetic field. It was observed that magnetic fields affect optical properties of certain materials so that when linearly polarized light is transmitted through the material in the direction of an applied magnetic field, the plane of polarization is rotated. This is the Faraday magneto-optic effect also referred to as the Faraday rotation, where the amount of rotation is proportional to the magnetic field H and the path length l. In the case of the MOI, the images are formed by distortions in the magnetic domains of the magneto-optic sensor in response to the external fields and are relatively insensitive to the strength of these fields. That is, the images are of a binary nature, which form with some minimum field strength. A schematic of the MOI instrument is shown in Figure 3. A foil carrying alternating current serves as the excitation source and induces eddy currents in a conducting test specimen. Under normal conditions, the associated magnetic flux is tangential to the specimen surface. Anomalies in the specimen generate a normal component of the magnetic flux density, which the magneto-optic sensor images. The sensor used in the instrument consists of a thin film of bismuth-doped iron garnet grown on a substrate of gadolinium gallium garnet. These films exhibit three important properties that are crucial for generating a magneto-optic image, namely, 1) They possess uniaxial magnetic anisotropy, i.e. they have an ‘easy’ axis of magnetization normal to the sensor surface and a ‘hard’ axis of magnetization in the plane of the sensor. 2) They possess ‘memory’, i.e. if the magnetization along the easy axis of magnetization is removed the film will retain most of the established magnetization. 3) Garnet films possess a relatively large specific Faraday rotation θf which can be in the range of 3 or 4 degrees per micron of material thickness. If linearly polarized light is incident normally on the sensor, the plane of polarization of light is rotated by an angle θ given approximately by → → → → θ ≈ θf (k • M)l/(|k ||M|) → Where k is the wave vector of the incident light, l is the sensor 18

The AMPTIAC Quarterly, Volume 6, Number 3

Figure 3. Schematic of the Magneto Optic Imaging System

Bias Coil Sensor

Lap Joint

Induction Foil

→

thickness, and M is the local state of magnetization of the sen→ sor. Note that M is always directed parallel to the ‘easy’ axis of magnetization. When the reflected light is viewed through the analyzer, local occurrence of normal magnetic flux is seen as a ‘dark’ or ‘light’ area in the magneto-optic image depending on the direction of magnetization. When the MOI was introduced in 1990, inspectors had to watch images on a monitor that had to be carried with them. In addition, a second scan of the inspection area was necessary, because the linear eddy-current induction method produced a null in the magnetic field in the same direction as that of the current induction. The original equipment had only one imager with a frequency range of 6.5 kHz to 100 kHz. High interest in corrosion detection spurred development of an imager utilizing an even lower eddy current induction frequency (to 1.5 kHz to permit deeper field penetration). Further development replaced the monitor with a head mounted display (see Figure 1), making the inspector completely mobile. In addition, the imager incorporated a rotating current scheme that eliminated the need for a second scan. Images now reflect a complete 360 degrees of the area being inspected. During the mid nineties, an MOI system directed toward the inspection of gas turbine engine parts was developed. The result was the MOI 307, which is significantly smaller and lighter and was able to operate at higher frequencies for engine materials such as stainless steel and titanium. The MOI 308 was introduced in 2000 (see Figure 2). This system replaces the MOI 303 and MOI 307 with a system incorporating both imagers with one single control unit. The new MOI 308 is a modular design approach, allowing inspectors to purchase as much or as little of the equipment necessary to accomplish the inspections required by various procedures. It is available in the following configurations: • The MOI 308/3 (MOI 303 imager) has a frequency range of 1.5 kHz to 100 kHz. • The MOI 308/7 (MOI 307 imager) has a total frequency range of 1.5 kHz to 200 kHz. This smaller imager has two different eddy-current induction attachments. The low frequency unit has a frequency range of 1.5 kHz to 50 kHz and the high frequency unit has a frequency range of 20 kHz to 200 kHz. They are interchangeable and snap on the imager. • The MOI 308/37 includes both the MOI 303 and MOI 307 imagers.

(a)

(b)

Figure 4. MOI 308/3 Image (a) and MOI 308/7 Image (b) of Surface Cracks

Uncracked spot welds (not visible) next to fasteners

Photograph of spot weld sample

Cracked spot welds next to fasteners

Figure 5. MOI Image of Spot Welds. The Larger Split Images are of Fasteners and Smaller Images are of Defective Spot Welds. Linear Eddy Current Induction Mode is Used.

The primary difference between the MOI 308/3 and the MOI 308/7 is in the size and weight of the imagers and the size of the imager field of view (FOV). Due to its larger FOV the MOI 308/3 allows for easier interpretation of subsurface corrosion and permits a larger number of rivets to be seen at once. The higher frequency range of the MOI 308/7 allows a wider range of applications that include lower conductivity materials such as titanium and stainless steel with improved resolution. 1

Probability of Detection

0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0

0

MOI 301 Lab

20

40 60 80 Crack length from rivet shank (mils)

EC Lab

MOI 301 Field

100

EC Field

120

MOI 303

Figure 6. POD Curves for the MOI on Fatigue Crack Samples

Its low weight and small size facilitate its use in restricted areas as well as general surface and subsurface inspections. Examples of MOI images for surface cracks using the 303 and 307 imagers are shown in Figure 4. The real-time imaging capability of the MOI permits rapid scanning of lap joints for the inspection of spot welds which was originally performed by tedious spot probe methods at the Air Force Logistic Centers (ALCs). An example is shown in Figure 5. POD Studies for Surface Cracks The capabilities of the MOI for detecting surface cracks have been evaluated in independent probability-of-detection (POD) studies. Shown in Figure 6 are POD curves obtained at the Airworthiness Assurance Nondestructive Inspection Validation Center (AANC) at Sandia National Laboratories using the MOI on a set of fatigue crack samples produced to evaluate various inspection techniques. The four curves to the right were obtained using the earlier MOI 301 and an eddy current based sliding probe. The results were published in a Materials Evaluation article[1]. The furthest curve on the left was obtained subsequently using the newer MOI 303. No data on the same samples have been obtained with the MOI 307 imager. However, similar results as for the MOI 303 are to be expected.

The AMPTIAC Quarterly, Volume 6, Number 3

19

Backside of skin

Corrosion

Chem-milled Step

Figure 7. MOI Image of Simulated Third Layer Crack MOI image from front side of skin Figure 8. MOI Image of Corrosion on Actual Aircraft Belly Skin

Peak Value of Flux Density (Gauss)

Subsurface Defects Since the MOI is also eddy current-based, the depth of penetration of electromagnetic waves into conducting inspection materials may be varied by adjusting the eddy current induction frequency. This permits the inspection of conducting materials for subsurface defects such as fatigue cracks and corrosion. Figure 7 shows the MOI image of a third layer-simulated crack. The crack is made with EDM (electrical-dischargemachining) creating a notch which measures 0.200" from the 5/32" rivet shank. Two sheets of .040" aluminum representing the outer skin and doubler cover the third layer .040" aluminum sheet. Behind the third layer is another .040" tear strap. The MOI induction frequency is 3 kHz. 2.5 Figure 8 shows the image of corrosion from an actual aircraft belly skin sample. The piece was cut from the 2 aircraft after the corrosion was detected by an MOI from the outside of the aircraft. The skin is .063" 1.5 thick with a step to .093". The worm-like structure seen in the images are the magnetic domains 1 in the magneto-optic sensor. They interact with the magnetic fields 0.5 from the defects to form the images. When the MOI is scanned over the inspection area, the domain structure 0 remains stationary while the defect 0 5 10 images move. This facilitates distin-

20

The AMPTIAC Quarterly, Volume 6, Number 3

guishing subtler defect images from the background. Currently, image-processing algorithms are being developed at Michigan State University to perform this function. Computational Analysis Support Since MOI technology is relatively new, its full capabilities have not been fully explored, but the use of numerical computational methods has been very useful in evaluating some of its capabilities. An Iowa State University computational nondestructive evaluation (CNDE) group under Prof. L. Udpa

Figure 9. Peak Value of Magnetic Flux Density as a Function of Corrosion Dome Height for the Single and Double Layer Geometry.

Single Layer

Two Layers (Current MOI)

15 20 25 Depth of Corrosion Dome (%)

30

35

40

Figure 10. Computational Grid for Dimpled Countersink with Circumferential Crack

Figure 11. Magnetic Field Distribution for Dimpled Countersink with Crack

Figure 12. MOI Image of Cracked Countersink (image on right)

(now at Michigan State University) has performed these computations. One exercise was to determine the effect of the second layer of a lap joint on the detectability of corrosion on the backside of the first layer. The computations[2] were conducted for corrosion domes of depths 10%, 20%, 30% and 40% of total thickness at the bottom of an aluminum plate. The model was then modified to study the effect of a second layer of aluminum in the geometry below the first layer. The presence of a second layer of aluminum - below the corrosion, provides an additional path for the induced eddy currents, thereby reducing the magnetic flux at the sensor placed above the first plate. The peak values of magnetic field in the presence of the second layer are therefore reduced by a factor of→ 2. Figure 9 shows that peak amplitudes of the magnetic field B z max are a function of height h of the corrosion dome for both single and double layer geometry. These results are extremely useful in guiding further development of the MOI for the detection of corrosion in lap joints. This can be accomplished by a combination of increased sensor sensitivity and/or increase in eddy current induction. Such improvements are being implemented. One additional example is the numerical verification of an inspection used at an Air Force ALC for the detection of circumferential cracks in dimpled countersinks under rivet heads. Originally, the observation was made that the usual symmetrical MOI image of some rivet sites was distorted using linear current induction. Upon visual inspection, a circumferential crack was detected in the countersink region. This effect was later modeled and verified numerically[3]. The computational region is shown in Figure 10. The computed magnetic field distribution is shown in Figure 11. The predicted MOI image can

be inferred from Figure 11 by taking a plane cut parallel to the y-z plane through the field distribution. The resulting cross section will be the MOI image. Due to the asymmetry of the field distribution, the resulting image will be asymmetric. The actual MOI image is shown in Figure 12. Procedures Procedures for using MOI have been written by airlines, maintenance facilities and the military. In particular, Boeing, Lockheed, Cessna and Bombardier have developed commercial inspection procedures and the military has written Technical Orders for KC-135s, B-52s, E-2s, P-3s, and C-141s. Gulfstream has issued a corrosion inspection procedure. A recently issued procedure for the MOI uses traditional eddy current spot probe procedures for the detection of hidden cracks in the skin along the chem-milled (doubler) edge on the back side of the outer skin. It was found that the MOI is significantly faster and easier to use. Boeing personnel estimate that using the MOI for the chem-mill crack inspections would be faster by at least a factor of four including verification of MOI-detected cracks with a spot probe. Actual on-plane tests proved even faster. The MOI permits the inspector to simultaneously locate the doubler edge as well as scan for subsurface cracks along the doubler edges from the front side of the skin. On the other hand, using a spot probe, the inspector would first have to trace out the doubler edge, mark the location on the surface and then use the spot probe to inspect for cracks along the marked edge. This would be a very tedious procedure. There have also been some creative uses of the instrument, not detailed in manufacturer’s procedures. Considerable success

The AMPTIAC Quarterly, Volume 6, Number 3

21

has been achieved in using the MOI for screening old 707s for conversion to JSTARS specifications for the Air Force. The MOI was used to rapidly identify aircraft unfit for use. This resulted in significant savings by avoiding unnecessary, expensive repairs. According to Bill Pember with NorthropGrumman in Melbourne, Florida, initially these aircraft were evaluated primarily through visual screening and documentation review. They often required tremendous rework and repair to bring them up to the required level of quality and structural integrity. The cost associated with this labor-intensive task was difficult to accept. The newly developed philosophy (1998) which utilizes MOI in the assessment of aging aircraft has reduced this problem and significant savings are now being realized.[4] R&D activities and other applications Developing improvements to the MOI system is an on-going activity. Many of the improvements have been made as upgrades to existing units to minimize equipment obsolescence. Current activities include an FAA-sponsored research program to improve the magneto-optic sensor’s sensitivity and to develop image processing techniques to improve and automate defect recognition. PRI has also cooperated with NASA under a Space Act Agreement. The intent is to develop a prototype reader for Data Matrix symbols (used for direct parts marking (DPM)) covered by paint and other coverings. Summary The MOI technology represents a relatively new application of magneto-optics to NDT. It retains many attributes associated with eddy current testing with obvious advantages of providing a visual display of the results using only analog methods. In this article, we have summarized the development of the MOI since its introduction to the NDT community a decade ago. Although the instrument is now widely used by both commercial and military installations to inspect for surface and subsurface defects, the route to general acceptance and use by the community has

been long and serendipitous. As yet, there is no well-defined process or criteria by which new technologies are introduced and accepted into the marketplace. The OEM on whose aircraft the instrument will be used must approve use of the instrument for specific inspections. If no general procedures exist, then each airline must receive specific approval for its use. The difficulty is the development of general criteria applicable to all possible inspections of a multiple variety of structural designs. For more information The preceding article describes an NDT technique aiding in the search for corrosion and other flaws on aging aircraft. While the article details specific materials-related attributes of this system’s design and implementation, this is just one of many techniques available in the large field of NDT The interested reader is encouraged to consult our colleagues at the Nondestructive Testing Information Analysis Center (NTIAC, one of 13 Defense Technical Information Center IACs like AMPTIAC) who specialize in this field. NTIAC can provide more information on NDE techniques, systems and standards, as well as links to other resources in the NDE community. NTIAC can be reached at www.ntiac.com, or 1-800NTIAC39. References [1] “The Validation Process as Applied to the MagnetoOptic/Eddy Current Imager (MOI),” Vanessa Brechling (Northwestern University) and Floyd Spencer (Sandia National Laboratories), Materials Evaluation, July, 1995, Vol. 53, No. 7, pp. 815-818. [2] “Finite-Element Predictions of MOI Performance for Application to Aging Aircraft Inspection” Lalita Udpa, in Center for NDE News, Iowa State University, Vol. 10, Issue 3, Winter 2000. [3] Private communications from Professor L. Udpa and Dr. L. Xuan of Michigan State University. [4] Private communications from Bill Pember, NorthropGrumman, Melbourne, FL.

A MPTIAC Help us update our mailing list 315 . 3 3 9 . 7 0 9 2 f a x 315 . 3 3 9 . 7 1 0 7 a m p t i a c @ i i t r i . o rg

P l e a s e i n c l u d e t h e 5 o r 6 d i g i t I D n u m b e r o n t h e m a i l i n g l a b e l o f e a c h s u b s c r i p t i o n yo u w i s h t o u p d a t e .

22

The AMPTIAC Quarterly, Volume 6, Number 3

AMPTIAC Directory Government Personnel T ECHNICAL M ANAGER /COTR Dr. Lewis E. Sloter II Staff Specialist, Materials & Structures ODUSD(S&T)/Weapons Systems 1777 North Kent St., Suite 9030 Arlington, VA 22209-2110 (703) 588-7418, Fax: (703) 588-7560 Email: [email protected]

A SSOCIATE COTR S CERAMICS, CERAMIC COMPOSITES Dr. S. Carlos Sanday Naval Research Laboratory 4555 Overlook Ave., S.W. Code 6303 Washington, DC 20375-5343 (202) 767-2264, Fax: (202) 404-8009 Email: [email protected]

DEFENSE TECHNICAL INFORMATION CENTER (DTIC) POC Melinda Rozga, DTIC-AI 8725 John J. Kingman Road, STE 0944 Ft. Belvoir, VA 22060-6218 (703) 767-9120, Fax: (703) 767-9119 Email: [email protected]

ORGANIC STRUCTURAL MATERIALS & ORGANIC MATRIX COMPOSITES Roger Griswold Division Chief US Air Force AFRL/MLS 2179 Twelfth St., Bldg. 652 Wright-Patterson AFB, OH 45433-7702 (937) 656-6052, Fax: (937) 255-2945 Email: [email protected]

ENVIRONMENTAL PROTECTION & SPECIAL FUNCTION MATERIALS Dr. James Murday Naval Research Laboratory 4555 Overlook Ave., S.W. Code 6100 Washington, DC 20375-5320 (202) 767-3026, Fax: (202) 404-7139 Email: [email protected]

IITRI Personnel D IRECTOR , AMPTIAC David Rose 201 Mill Street Rome, NY 13440-6916 (315) 339-7023, Fax: (315) 339-7107 Email: [email protected]

D EPUT Y D IRECTOR , AMPTIAC Christian E. Grethlein, P.E. 201 Mill Street Rome, NY 13440-6916 (315)-339-7009, Fax: (315) 339-7107 Email: [email protected]

T ECHNICAL I NQUIRY S ERVICES M ANAGER David Brumbaugh 201 Mill Street Rome, NY 13440-6916 (315) 339-7113, Fax: (315) 339-7107 Email: [email protected]

Overworked? Overloaded? Could you use some materials engineering help? AMPTIAC can provide answers to materials-related technical questions. Here’s how it works: You contact our inquiry manager with the problem. The inquiry manager discusses the problem with you to make sure we understand exactly what you need. He then assigns the task to an AMPTIAC technical expert with knowledge and experience of the discipline in question. AMPTIAC maintains the DOD’s knowledge base in advanced materials; nearly 220,000 technical reports addressing all classes of materials. Our database contains information on properties, durability, applications, processes, and more. Plus if we don’t find it in our resources, we also have direct access to NASA and DOE databases. With this tremendous amount of data, our engineering staff can save you time and money by quickly providing you with off-the-shelf information or technical solutions that directly meet your needs. For smaller inquiries and bibliographic searches we can provide you some information free of charge. Larger efforts are on a cost-reimbursable basis but under no circumstances do we begin work before you accept our quote and issue us a purchase order. For more information on how we can help you, please contact AMPTIAC’s Inquiry Services Manager, Mr. David Brumbaugh, at (315) 339-7113. A Recent Example A government contractor asked us to locate and compile properties of carbon fiber composites at cryogenic temperature. In less than 2 weeks we: • performed a literature review • identified 162 relevant technical reports • reviewed the reports and extracted appropriate data • organized 152 pages of data in a binder • and cross-linked the data to a searchable spreadsheet The resultant data book provided the contractor with valuable information they needed in their effort to design a satellite structure.

The AMPTIAC Quarterly, Volume 6, Number 3

23

QUARTERLY MaterialEASE: Tips and Pitfalls of Corrosion Testing MOI System Allows Examiners to “See” Defects in Metal Materials Data Exchange Made Easier on the Web … and more!

Update our mailing list... IIT Research Institute/AMPTIAC 201 Mill Street Rome, NY 13440-6916

see page 22 Non-Profit Org US Postage Paid Utica, NY Permit No. 566