materials for a non-steady-state world...2014 distinguished lectureship in materials and society...

2014 Distinguished Lectureship in Materials and Society

Materials for a Non-Steady-State World

DOI: 10.1007/s11661-016-3442-6� The Minerals, Metals & Materials Society and ASM International 2016

I. INTRODUCTION

THIS is the 44th lecture in the series of ‘‘Materials andSociety.’’ The first lecture was presented in 1971 by Prof.Harvey Brooks, Harvard University, and was titled ‘‘Ma-terials in a Steady-State World.’’[1] The focus of ProfessorBrook’s remarks was on sustainability and economical useof materials in design with the thought that a steady statewould eventually be reached between consumption ofmaterials by industrial societies and the availability ofmineral resources. He posited that technical progress inthree areas would be necessary in order to attain thisdesired state of affairs, anticipating that the world wouldcontinue to industrialize and thus consume ever moreresources: (i)More ore bodieswould be discovered, and theextraction fraction of metals from them would be

improved; (ii) Substitute materials would be developed toreplace those that become scarce; and (iii) New technolo-gies would improve the recycling of materials.The title of this talk, ‘‘Materials for a Non-Steady-

State World,’’ is aimed at a different theme: expectationsfor continued improvement in the functionality andquality of products continue to rise non-linearly as theeffects of progress are observed and experienced. Sincematerials are an essential ingredient of all physicalproducts, the challenge to the materials community is tocontinue to develop and implement improved materialssolutions that directly impact the quality of life at amuch faster pace than a typical 10-year timeline in orderto better align with the typical 3- to 5-year productdevelopment timeline. Furthermore, societal expecta-tions are that the improved material solutions willmeet all requirements so that performance of the endproduct is maintained. Importantly, accelerated materi-als development can offer an attractive value proposi-tion for the customer and the manufacturer.These considerations have been a driving force for the

US government’s Materials Genome Initiative.[2] TheMaterials Genome Initiative (MGI) is a multi-agencyinitiative designed to create a new era of policy,resources, and infrastructure in the effort to discover,produce, and deploy advanced materials twice as fast. Inline with MGI, Integrated Computational MaterialsEngineering (ICME) has the vision of integratinginformation across length and time scales for all relevantmaterials phenomena and enables concurrent analysis ofmanufacturing, design, and materials for a complete endproduct.[3,4]

Much effort has been expended in the developmentand verification/validation of computational models,and supporting databases that can estimate materials

ROBERT E. SCHAFRIK

ROBERT E. SCHAFRIK, Sr., formerly General Manager with theMaterials and Process Engineering Department, GE Aviation, Even-dale, OH, is now retired. Contact e-mail: robert.schafrik @yahoo.com

Robert E. Schafrik, Sr. retired from GE Aviation in April 2014 afterserving for 17 years as general manager of Materials and ProcessEngineering. This department develops and qualifies the materials usedin GE’s aero-engines and their land and marine derivatives. Prior toGE, for seven years he was staff director for the National MaterialsAdvisory Board at the National Research Council (NRC). Prior tothis, Bob was vice president of research and development forTechnology Assessment and Transfer, Inc. for three years. He servedin the U.S. Air Force for 20 years in a variety of R&D and advancedaerospace system acquisition capacities. Bob chaired the NRC’sNational Materials and Manufacturing Board 2012–2015. He is aFellow of ASM International and is a member of the NationalAcademy of Engineering. He received ASM International’s WilliamHunt Eisenman Award and GE Corporate’s Edison Award. Bob has aPh.D. in metallurgical engineering from Ohio State University, a M.S.in Information Systems from George Mason University, a M.S. inAerospace Engineering from the Air Force Institute of Technology,and a B.S.in Metallurgy from Case Western Reserve University.

Article published online March 29, 2016

METALLURGICAL AND MATERIALS TRANSACTIONS A VOLUME 47A, JUNE 2016—2539

http://crossmark.crossref.org/dialog/?doi=10.1007/s11661-016-3442-6&domain=pdfhttp://crossmark.crossref.org/dialog/?doi=10.1007/s11661-016-3442-6&domain=pdf

properties, and can assist in the development of optimalmaterials processing routes. Many presentations atmaterials conferences address some aspect of ICMEand the MGI. Indeed, there are entire technical confer-ences now dedicated to both topics.

There are three strategic elements that are necessaryto achieve more rapid commercial application of newmaterials in addition to computational tools. Considerthat even if computational models were perfect, reducedimplementation time is not a given. These three addi-tional elements can significantly reduce the time andeffort required to insert new materials into products:

1. Forward-looking materials strategies that contain afeasible menu of materials advancements that couldadd significant value to an application of interest.

2. Materials development process with a strong appli-cation focus so that the technology is championed bythose responsible for materials implementation deci-sions. Development and implementation cannot beperformed as completely separate tasks for anaccelerated program.

3. Creation and fostering of high-performance teams toexecute materials development.

For simplicity, the term ‘‘material’’ is used instead of‘‘material system.’’ In reality, an integrated materialsystem must be developed for a particular application;this includes the base material, the necessary environ-mental and tribological coatings, the suite of processesemployed to produce and inspect the material, andmethods of joining with other structural elements. Also,to avoid redundant wording, the term ‘‘material’’ is alsomeant to include new process developments, such asadditive manufacturing.

II. DRIVING FORCES PROMOTING ANDTEMPERING TECHNOLOGY CHANGE

The primary impetus for change in materials technol-ogy is responding to the competitive landscape; amanufacturer can enhance its position in the marketplaceby offering a product that is superior to those of thecompetition. Design changes are desired that cost-effec-tively increase customer value, such as greater perfor-mance, better safety and reliability, lower price, andreduced maintenance burden. Design changes could alsoresult in reduced manufacturing cost so that the manu-facturer’s operating margin can be increased, freeing upresources for additional product development.

New designs, or changes to a current design, have therisk that not all the changes will perform as intended,with the consequence that the competitive positioncould worsen. Thus, understanding the possible risksand the resources required for risk mitigation strategiescan temper the go-forward decision for a new material.Some of the risks that are typically considered includethe following:

� Materials and process development may not becompleted on time; e.g., difficulty in scaling up toproduce full-scale components from subscale experi-

ence. These problems are usually resolved over time,but the lament that one ‘‘cannot invent on a timeline’’is unfortunately true, so that the completion datecould be uncertain.

� Not all critical requirements may be satisfied by thenew material, causing a debit in performance, andpossibly a redesign of the system to accommodate amaterial property shortfall. For instance, for applica-tions outside of past experience, additional criticalmaterials requirements that were not previously fore-seen may be discovered during testing or early in-ser-vice operation. Another potential cause of a propertyshortfall is not having full understanding of the processcontrol limitations for full-scale processing.

� Accelerated tests may not properly simulate long-term materials performance in the actual serviceenvironment.

� The supply chain may require additional time to scaleup for production quantities that meet the necessarydelivery rate, quality, and cost metrics.

The conventional timeline for introducing a newmaterial is often stretched out to deal with risk items.The development timeline can be divided into fourcases.[5] In this context, a critical material application isone in which a material failure would cause significantloss in functionality such that the end product would notwork properly, and have to be repaired or replaced.

I. Modification of an existing material or process foruse in a non-critical application—2 to 5 years.

II. Modification of an existing material or process foruse in a critical application, such as a critical struc-tural component—4 to 6 years.

III. Insertion of a new material into a critical applica-tion—10 to 12 years.

IV. Introduction of a new material class or a new pro-cessing route—up to 20 years and beyond.

As will be described in subsequent sections, for CasesI, II, and III a 50 pct reduction in the development timefor insertion into an application is potentially possible atrelatively low risk. Cases I and II would have asubstantial amount of information about the material,the supply chain would already exist, and the businesscase for insertion could be developed with confidence.For case III, a great deal of data will exist; it may notinclude full-scale production though, so the risk ofinsertion will be higher. Applying a new material in anexisting application is lower risk since the requirementsthat the material must meet are known. Inserting a newmaterial into a newly designed application is morechallenging since all the requirements may not be knownuntil product testing is accomplished, and the ability ofthe production process to achieve the requirements ofthe fully designed components can be assessed. This isespecially true of highly complex systems, such as jetengines, in which all the design parameters cannot beaccurately predicted, with the consequence that thematerial properties may have to be adjusted as a resultof what is learned during product development testing.[6]

Case IV involves the development and application ofa new material class, such as an intermetallic structural

2540—VOLUME 47A, JUNE 2016 METALLURGICAL AND MATERIALS TRANSACTIONS A

material, or a completely new processing route, such asadditive manufacturing. There are multiple options thatmust be investigated, and multiple application oppor-tunities that must be sorted out. So a 50 pct reduction indevelopment time would be difficult to achieve unless acostly project is launched to concurrently exploremultiple options.

The titanium aluminides are an example of Case IVthat took longer than 20 years to be applied. (Thedevelopment time would have been much longer withoutthe pioneering work accomplished by the Air ForceResearch Laboratory’s Materials and ManufacturingDirectorate, the Office of Naval Research, and theStructures and Materials Division at the NASA GlennResearch Center.) These materials had the attractivepromise of replacing nickel-based superalloys in certainapplications without loss in performance at half theweight. On the other hand, they also possess much lessthan half the ductility at room temperature, so newdesign rules had to be created to avoid the possibility ofcracking during processing, installation, and operationat lower temperatures. Plus the supply chain had to beestablished, from the synthesis of the material to specificprocessing steps. Material and process specificationswere developed and implemented such that the newmaterial could be consistently produced to technical andcost requirements.

GE began its investigation of the titanium aluminidesin the 1980s. US patent 4916028A was granted in 1990for a chemistry that provided a balance of mechanicalstrength and ductility/fracture toughness. The alloy thatbest survived the gauntlet of development tests wasTi-48Al-2Nb-2Cr. Many potential applications for TiAlin a jet engine were considered. GE focused on the largeturbine blades since the weight savings would beconsiderable; this translates into improved fuel effi-ciency. A new design approach was developed toincorporate fracture mechanics analyses beyond whatis typically done for conventional nickel-based superal-loy turbine blades. Also, the casting suppliers werechallenged to establish TiAl turbine blade castingprocesses capable of meeting rate production at thenecessary quality levels. The raw material suppliers hadto scale up from providing a few thousand kilograms ofingot material per year to 450,000 kg per year.[7]

Figure 1 shows the cast TiAl, GE 48-2-2, turbineblade uninstalled and installed in the last stage of thelow pressure turbine rotor of a GEnx engine. This is thefirst FAA certified structural intermetallic materialapplication. The figure also lists the system-level certi-fication requirements that the TiAl turbine blade satis-fied in gaining certification. It is challenging to convertthese system-level requirements into specific materialproperties, so these certification tests can result insurprising results. The TiAl turbine blade past thesetests without difficulty.[7]

The TiAl low pressure turbine blades have performedflawlessly in-service. The Boeing 747-8 entered revenueservice in 2011; it is powered by four GEnx-2B enginesthat contain one stage of TiAl turbine blades. TheBoeing 787-8 entered revenue service in 2012; it ispowered by two GEnx-1B engines, each of which

contains two stages of TiAl turbine blades. Conse-quently, each airplane model flies with 400 TiAl turbineblades, As of the fall of 2014, 100,000 TiAl turbineblades had entered service in 175 aircraft encompassing500 engines; altogether these turbine blades had accrued2.5 million hours without failures. So the long, drawn-out materials development process does work. Anaccelerated development program would have allowedbenefits to be realized much sooner if the monetizationof the benefits was greater than the development andscale-up costs while considering the probability ofsuccess for a first application of a new class of materials.A reasonable goal for Case IV materials and processes

would be reducing the development time by half oncethe material chemistry/microstructure/architecture hasbeen well developed and characterized, or a new processhas matured to the point of consistently producing adesired output, such that an explicit application can betargeted.

III. STRATEGIC FOUNDATIONS FORACCELERATING R&D AND DEPLOYMENT

OF NEW MATERIALS

Over the course of the past decade, GE Aviation hasintroduced a large number of new materials andprocesses as it refreshed its family of aero-turbineengines and responded to demands of the marketplace.The new materials and process technologies included thefollowing:

� TiAl for cast turbine blades.� Advanced high temperature superalloy for single

crystal high pressure turbine blade alloy.� Advanced high temperature superalloy powder alloy

for turbine disks.� Ceramic matrix composite, SiC/SiC, for turbine

shrouds.� Additive Manufacturing to produce advanced turbine

combustor fuel nozzles.� Polymer composite flowpath spacer, designed to

‘‘rotating part standards’’; i.e., same standards asused for turbine disk design.

� Light weight carbon fiber composite for fan cases.� Advanced tribological coatings for compressor vanes

and blades.� GE1014 ultra high torsional strength steel for fan

shafts.� NiAl for turbine blade coating.� Advanced solid state joining for titanium and super-

alloy components.� Low rhenium content, René N515 alloy for single

crystal turbine blades.� High temperature cast and wrought alloy, René 65,

for turbine rotating and static structure applications.

This significant development activity provided oppor-tunity to hone the approach to insert new materials intoapplications. There were four necessary elements: mod-eling and simulation of the material and processing,forward-looking materials strategy, robust managementof the materials development process, and formation of


high-performance development teams. Since much isbeing written about modeling and simulation, theremainder of this paper will focus on the last threeelements.

The accelerated programs, undertaken to meet acritical business need, were distinguished from thoseconducted by the standard development process, by thedegree of concurrency in the tasks, rather than merelyperforming each sequential task faster. Thus the crux ofthe accelerated process was to direct the concurrencywith a deliberative, thoughtful method.

A. Forward-Looking Materials Strategy

At the strategic level, aerospace applications contin-ually demand new materials that are lighter weight,possess higher operating temperature capability, andhave increased durability in the service environment.And not only must a new material excel in at least one ofthe desired properties, it must also have a practicalbalance of all the other properties required for anapplication. Therefore, establishing a well-definedmethod for highlighting and assessing promising newmaterials and process directions is important for build-ing a storehouse of practical advanced materials tech-nology ideas.

The first step in constructing a storehouse ofadvanced materials ideas is to simply list the high levelmaterials and process technologies of interest in theapplication domain. For aerospace, this list wouldtypically include nickel-based superalloys, titanium,high strength steels, polymer matrix composites, ceramicmatrix composites, anti-corrosion coatings, additivemanufacturing, solid state welding, etc. It is usuallybest to organize these technologies by material orprocess type rather than by application, such as ‘‘nextgeneration turbine blade’’; normally a design engineerselects the material for an application, and one of thegoals of materials experts is to provide options for thedesigner. For instance, a turbine blade could be madefrom a superalloy, a titanium aluminide, or a ceramicmatrix composite, each of which would have a separatematerials development strategy. For strategies focusedon processes, such as solid state joining and additivemanufacturing, multiple materials would be included.For example inertia welding of titanium and nickelalloys could be involved in assessing new directions forsolid state joining.

For each category of technology, the current state ofthe art is assessed, including current known researchdirections, as well as relevant intellectual property(IP)—licensing technology of interest could save con-siderable time and resources. Then materials experts,together with other experts, use their inventive genius toexplore new technology directions, potential benefitsand anticipated challenges. It is important to understandthe entitlement level for the materials and processes ofinterest, which sets an expectation regarding the point ofdiminishing returns. For example, highly alloying asuperalloy can not only improve strength, but it can alsoincrease the tendency for macro-segregation, lowerincipient melting point, and increased density.

An important step in the process is to identify futurepriority applications for each technology idea, togetherwith the most important requirements that are pro-vided by design engineers and customer requests.Understanding the gaps between the currently avail-able material solution and the potential additionalcapability offered by a new material is essential toconstruct a program plan that demonstrates feasibilityand provides necessary inputs into the value caseanalysis. This usually requires accomplishing a prelim-inary design of a specific component using anticipatedmaterial properties. Prediction of properties usingcomputational tools, supplemented by selected smalllaboratory-scale testing, can add to the confidence levelof what properties could be achieved. If a designengineer concludes that the estimated properties are ofinterest, a development plan is outlined. It includespreliminary work to validate the projections, and aroadmap for full development that leads to insertion ina specific application. The roadmap could include apartnering strategy.

B. Robust Management of Materials Development

The opportunities highlighted in the materials strat-egy plan that proceed into the next phase of develop-ment require the agreement by the design organizationand the engine line business manager. The primaryconsiderations are as follows: the materials technologywould add significant value to a specific application, andthe development cost and risk are manageable withinthe expected timeline. The targeted application could bein an existing aero-engine or one that is slated fordevelopment. Those opportunities not selected forscale-up remain on the materials strategy list for furtherinvestigation, except for those for which the determina-tion is that there no potential application in theforeseeable future.There are three phases to materials development: (i)

Establishment of Feasibility, (ii) Demonstration ofApplicability, and (iii) Maturation and Scale-up. Thedevelopment time is measured from the start of Feasi-bility to the insertion of materials or process technologyat the end of the Maturation phase. Milestones thatmeasure progress follow the framework of the Technol-ogy Readiness Levels (TRL) and Manufacturing Readi-ness Levels (MRL).[8,9] Accelerated programs have adegree in concurrency for these phases, while standardprograms do each phase sequentially. A schematic ofconcurrency is depicted in Figure 2.Each development program establishes the level of

concurrency appropriate for that program. Importantfactors for determining the appropriate degree ofconcurrency include the following: an assessment ofthe benefit level if the development is successful, thedegree of risk for doing some tasks concurrently, andthe desired insertion date for the technology. In the past,the materials development timeline was somewhatdisconnected from the product development timeline;materials development took much longer than productdevelopment, and consequently many opportunities toinsert new materials and processes went astray.[10]


Since the design engineer selects the materials andprocesses to implement the design, a decision was madeat GE to involve the design engineer in the leadership ofthe materials development program from the beginning.Previously, the design engineers were briefed as keymilestones were achieved. This process did not provide asufficient amount of interchange between the designteam and the materials development team for anaccelerated program. Design requirements evolve, mate-rials property data are constantly being generated,materials composition can change slightly with time,etc. Having the lead design engineer as part of theleadership team eliminated communication gaps, andled to much smoother insertion of the materialstechnology.

The Feasibility Phase has the objective of demon-strating the feasibility or suitability of a new material orprocess for a selected application. During this phase, thefollowing main activities are accomplished:

� The technology concept is refined using computa-tional tools and the knowledgebase of experts. Thisincludes developing the full suite of requirements forapplications of interest in partnership with designengineering. Significant challenges are explicitlyidentified, along with mitigation strategies.

� Initial feasibility of the technology is confirmed,usually through a combination of further analysesand proof-of-concept experiments. The test results arecompared to analytical predictions to gauge theaccuracy of the predictions and provide insight intoareas for further improving the prediction tools andmethodology.

� Manufacturing methods and production cost are as-sessed in order to validate the value case for theproject.

The Demonstration Phase generates data to charac-terize the risks, costs, and benefits on inserting thetechnology into a particular application. The primarygoal is that the product decision-makers will havesufficient information to make informed decisionregarding proceeding with full-scale development. Themain activities during this phase include as follows:

� Demonstrate capabilities in a portfolio of laboratorytests, component rig tests, and engine tests. Completea design incorporating the material and validate therequirements. Manufacture components using pro-duction-scale processes.

� Develop manufacturing processes and procedures.Identify long lead items and key supply chain ele-ments.

The Maturation Phase incorporates the new materialinto the design of a particular component, and producesthe component using full-scale production equipment.The following activities occur during this phase:

� Generate a complete set of material property datausing specimens extracted from production hardware.

� Establish the processing details, including specifica-tions and standards.

� Qualify the material, and certify the product appli-cation with the appropriate aviation authority, suchas the Federal Aviation Administration (FAA).

� Monitor in-service performance after implementationto verify that the material meets expectations. If all

Fig. 1—Gamma TiAl turbine blade and certification requirements.

Fig. 2—Concurrency of material development phases.


requirements are not being met, a development effortcan be targeted at providing the necessary upgrade.

Figure 3 shows that the product creation cycleemploys the same three phases as the technologycreation development cycle: Feasibility, Demonstration,and Maturation. In this depiction, the technologycreation phases are sequential; however, the phaseswould have concurrency for an accelerated program. Acardinal rule of product development is that onlymature technologies from the materials Maturationphase should be considered for insertion into theproduct’s Demonstration phase. Thus, materials tech-nology must necessarily lead the product creation cycletogether with substantial interaction with the productcreation team. This reality is a fundamental reason thathaving a well-prepared materials strategy is important.

C. Formation of High-Performance Development Teams

Development programs necessarily involve a greatdeal of interaction between people with different exper-tises, different skill levels, and different roles. A pas-sionate, knowledgeable, small, highly focusedhigh-performing team is crucial to the success for anydevelopment effort, especially for an accelerated pro-gram. Many different materials, manufacturing, systemengineering, and design specialists must work togetherseamlessly toward a challenging common objectiveunder a defined process that requires each team memberto contribute complementary expertise. The team leaderand senior business leaders have the crucial role ofremoving obstacles for the team, providing necessaryresources, communicating the importance of progresstoward the goal, and holding the team accountable forthe results. During the course of a development pro-gram, many issues are worked across the disciplines.Usually these issues are not confined to a singlediscipline, but have multiple interactions. The team’sgoal is to quickly and constructively address theseissues.[11] Upon completion of the objective, the team isdisbanded and the people re-assigned to other projects.

Attracting people to serve with enthusiasm on theseteams has not been difficult. Consider that there arethree behavioral realities that affect the behavior ofpeople: historical reality (cultural background, educa-tion, training, inherent personality, etc.); current reality(personal gain, peer pressure, habit, loyalty, expecta-tions, optimism, etc.): and anticipatory reality (dreams,visions, utopian ideas). Of these three realities, antici-patory reality is by far the most important, with morethan twice the combined influence of historical realityand current reality.[12] Anticipatory reality is for-ward-looking. Most people gain great satisfaction inworking on technology that is successfully implementedwithin a reasonable amount of time. In the past, with thelong development cycle and uncertainty in the endapplication, a materials developer could specialize on asingle technology for an entire career, such as titaniumaluminides, with the hope that the technology wouldeventually be selected for an engine application. Underthe new approach, the technology is targeted at an

application early in the development cycle in partnershipwith design engineering. Thus, people skilled in mate-rials science, development, and application now have theexpectation that the new materials technology could beinserted into an application within a few years. Thisallows the opportunity to work on multiple materialstechnology developments over a career. Such a prospectis very attractive to many materials scientists andengineers, as well to other team members.

IV. ACCELERATED MATERIALSDEVELOPMENT PROGRAMS

In recent years, GE Aviation has conducted severalaccelerated materials development programs, using themethods described in this paper. Two of the programsare discussed below. Both of these examples replacedexisting materials in production engines; the time fromstart of the development program to engine certificationby the FAA was less than half the traditional time.The first example is a new low rhenium content

single-crystal turbine blade alloy, René N515. Themotivation was a strategic concern on the long termavailability of rhenium. The second example is a newhigh-temperature cast and wrought turbine disk alloy,René 65, that could replace a powder metallurgy diskalloy that had to be iso-thermally forged; with the largeincrease in engine sales, a look-ahead in iso-forgeindustrial capacity caused concern that major equip-ment investments would have to be made in the nearfuture to avid constraining disk production capacity.

A. Low Rhenium Content Turbine Blade

Rhenium has been found to greatly strengthennickel-based superalloys at high temperature, and assuch has been a widely used alloying addition togeneration two single-crystal superalloy blades that areemployed in the first stage of the high-pressure turbine.Rhenium is very rare metal. There are no direct oredeposits of rhenium; it is produced as a byproductduring copper and molybdenum refining. For more thana decade, its price was less than $2000/kg. In 2006 itsprice doubled; by 2008 the price had increased to fivetimes the baseline amount. GE Aviation did an analysisto determine the root cause of the price increase. It wasconcluded that the usage of rhenium in turbine bladesuperalloys was approaching the yearly productionvolume of the element—so the price increase wassupply–demand driven. Furthermore, extrapolation ofthe rhenium usage trend line indicated that GE’s use ofrhenium alone would exceed the world’s availablesupply within a few years if no actions were taken.Consequentially, a major program was instituted

immediately to work with customers and suppliers torecycle all rhenium-containing alloys. Another actionwas to substitute other alloys in applications that didnot require the strength of the rhenium-containingalloys (previously, standardizing on a few alloys wasdesired by the supply chain). Finally an assessment wasmade regarding the development of a new alloy


containing half the rhenium of the then standard GERené N5 alloy without compromising any capability;i.e., provide a form-fit-function material substitution.

During the materials strategy process for superalloyturbine blade materials, digital tools had been developedto predict turbine blade properties based on chemistry.These tools relied on a wealth of data from thousands ofturbine blade tests of alloys with various chemistriesperformed by GE over a 25-year time period. Thesetools were employed to digitally evaluate chemistryiterations based on a design-of-experiments (DOE)analysis. The tools estimated strength, phase stability,and propensity to form undesirable phases. At the sametime, several chemistries were selected to be cast into testspecimens. Over a 3-month time period, the predicteddata were able to be compared to the test data; there wasreasonably agreement. These results added confidencethat a new turbine alloy could be developed to replacethe René N5 alloy with half the rhenium content(1.5 wt pct) with essentially the same properties as RenéN5. Figure 4 shows the principal high-level require-ments that the new turbine blade alloy had to meet.[13]

During the subsequent 3 months, the models wereexercised to demonstrate that it was possible to movedirectly from small heats to full-scale heats, skipping thetime and cost of subscale heats. Full-scale heats of thematerial were ordered so that there was sufficientmaterial to make a plethora of test specimens, and toinitiate trial castings of production turbine blades.Ascertaining the castability of the new alloy wasessential since redesigning all the casting tooling acrossall the engine lines would have taken considerable timeand effort. Also, production processes such as machin-ing, grinding, and coating were assessed on full-scaleproduction blades. A few blades were inserted into aground-test development engine to further assess per-formance under actual operating conditions.

Two years from the initial start of the program, thenew alloy, known as René N515, was certified by theFAA, after engine testing in multiple engine lines. The

traditional time required to develop a new turbine bladealloy was 4 to 6 years, and even longer in some cases.René N515 was incorporated into multiple applicationsacross all the GE family of engines without requiringredesign of the turbine blade; slight adjustments in thecasting tooling were required in some instances to avoidrecrystallation defects. Figure 5 depicts the overall RenéN515 development timeline. The critical decision wasauthorizing full-scale heats of the material after the6-month analysis period.[14]

René N515 is currently used in all high-performanceGE Aviation engines. Each engine model has differenttemperatures and different temperature profiles. Themechanical properties of René N515 at all the differenttemperature conditions were slightly different than RenéN5 (it is slightly stronger than René N5 at sometemperatures, and slightly weaker at some tempera-tures). A decision was made to assign a system engineerfrom the advanced design organization as co-leader ofthe development program to work with the designengineers from each engine model to continually

Fig. 3—Technology (material) and product development stages.

Fig. 4—Required material capabilities for alloy René N515.


evaluate the ability of René N515 to be a form, fit, andfunction replacement for N5 as the results from theextensive mechanical test program were completed. Thisorganizational decision saved considerable program-matic time as the design engineers did not have to waituntil milestone reviews to see the latest data andsubsequently perform their analyses. The system engi-neer also sets the priorities for the sequence of enginemodels conversion to René N515.

The René N515 blades have performed comparable toRené N5 blades in service. The price of rhenium hassteadily declined as the usage of René N5 has declined;in January 2016 the price of rhenium was under $3000/kg.[15]

B. High Temperature Cast and Wrought Alloy, René 65

Alloy 718, a cast and wrought alloy, is the mostwidely used material for turbine disks. However, it haslimited temperature capability. As the performance andtemperatures of aero-turbine engines have increased,alloy 718 did not have suitable mechanical strength as afirst-stage turbine disk material. More highly alloyedsuperalloys with greater temperature capability wereinvented using powder metallurgy techniques. Disksmade from this material require additional processingsteps and thus are more costly: molten metal is atomizedinto powder; the powder metal is loaded into a can andthen usually hot iso-statically pressed (HIP) andextruded to fully consolidate the material; and finallythe disk is forged to shape from a slab cut from theextrusion in a special press that uses a slow deformationrate at an iso-temperature.

With the increase in the production volumeof aero-engi-nes, GE Aviation recognized that the suppliers would nothave sufficient iso-forge capacity to produce all the neededdisks within 5 to 10 years unless suppliers made majorinvestments to acquire additional iso-forges, which costmillions of dollars. Since many of the applications of thepowder alloy René 88DT (R88DT) did not need the fulltemperature capability of that alloy, the challenge was tore-examine producing a cast and wrought (C&W) versionof R88DT that could replace at least half of the R88DTapplications. This had been tried three times previouslyover the past 20 years. Those efforts were unsuccessful.Other alternative alloys were considered; however, nonemet all the strength requirements.

The latest computational tools were applied toevaluate chemistry changes to R88DT that would make

it processable by C&W, while still having sufficienthigh-temperature strength to replace some applicationswhere R88DT was being used. Process and propertymodeling indicated that there was a narrow window thatcould be successful for the chemistry of interest. Adecision was made to not use subscale trials—to goimmediately to full scale. GE Aviation partnered withmaterial supplier, ATI Allvac for this effort. ATI hadmade major investments in new furnace capability and acomputer-controlled ingot breakdown facility that werecapable of meeting the narrow processing window. Thenew alloy, René 65 (R65), was successfully cast andwrought. The material was then forged on conventionalhot die presses to make full-sized disks. R65 is also beingevaluated for large static structure applications.R65 was conducted as an accelerated program. It was

successfully completed in less than half the time that atypical turbine disk program has taken—the traditionaltimeline for a new disk alloy is 10 to 12 years; R65development took 4 years from start to FAA qualifica-tion. Since it was aimed at applications in multipleengines, an unprecedented amount of production veri-fication was undertaken. A typical disk developmenteffort uses one or two forging suppliers; the R65program used 10 forging suppliers. A total of 250forgings were produced, covering the gamut from thesmall CF34 engine to the very large GE90 engine. Atotal of 4500 mechanical property tests were performed,consuming 1 million test hours.[16]

As was the case for the René N515 accelerateddevelopment program, there was a significant concur-rency in the tasks for the R65 development program. Ahigh-level development timeline is depicted inFigure 6.[17] Since the team had high confidence in theselected chemistry, the majority of activity was focusedon developing processing steps for making componentsof different sizes and shapes, and on mechanical testingusing test specimens extracted from hardware producedusing full-scale production processes.GE Aviation has licensed the material to several other

engine manufactures in order to build industrial capac-ity. In 2014, usage was approaching 450,000 kg per year.

V. GO-FORWARD CHALLENGES

The success with accelerated materials developmentprograms demonstrates that reducing the timeline byhalf is possible. Experience to date points to twoopportunity areas for further improvement in thedevelopment process: formation of high-performanceteams, and continuing to improve the cyber computa-tional tools and infrastructure.The first challenge relates to further understanding

and molding of high-performance teams. These teamsare formed to achieve development program goals thathave high impact. The team members really enjoy theirexperience serving on these teams, but sometimes aredisappointed when the next team they join is not ashigh-performance. So the challenge is to have everyteam be high-performance, whether working on a highvisibility development program or some otherFig. 5—René N515 development timeline.


important, less visible task. This is very much aleadership challenge since the ingredients of institutinghigh-performance teams are known. Success with theseteams builds additional successes. Over time, this willresult in a cultural transformation in the workplace.

The second challenge addresses the need to continueto improve the digital tools and supporting infrastruc-ture used by materials scientists and engineers. The useof modeling tools has increased year-over-year, andalmost every materials development effort has madeproductive use of multiple digital tools.

There has been a non-linear increase in the numberand use of digital models over the past 30 years. Earlymodels focused on process modeling, such as forging,casting and heat treatment. Models were employed tohelp solve problems. Even though many tools used2-dimensional finite elements, they provided usefulinsights. Soon models began to appear that predictedmicrostructure evolution based on thermodynamicknowledge. Over the past 10 years, models that predictproperties, such as tensile, fatigue, and creep, have beendeveloped. Some models employ heuristics while othersare based on fundamental science. In addition, somecodes are using 3-dimensional voxels to improve theaccuracy of the analyses. To a large extent, the expo-nential increase in computer performance has madethese improved models feasible for the materials com-munity. Another important trend is the linking ofmodels so that complex questions can be addressedwithout requiring significant human intervention. Thecomposite materials area has been particularly effectivein integrating models. As these models have improvedand become more numerous, they have contributedsignificantly to the ability of shortening the materialsdevelopment timeline without increasing the cost andrisk of the development.

Suggestions of further improvements in digital toolsinclude:

� Continue to improve computational models based onfundamental understanding of the phenomena beingmodeled, including databases that support the mod-

els. Priority would be models with the followingcapabilities:

� Predicting microstructure evolution during pro-cessing.

� Predicting the effect of defects and microstructurevariation on mechanical properties.

� Scaling mechanical properties obtained from sub-scale laboratory tests to those expected for full-sized components.

� Estimating the impact on mechanical properties asa result of the interaction of different degradationmodes, such as corrosion, fatigue, and creep.

� Standard framework for linking tools together so thatoptimizations could readily be performed.

� Smooth integration of materials predictive modelswith design analysis tools.

A cyber infrastructure is needed that creates alow-maintenance collaborative materials ecosystem.Such an ecosystem, as notionally depicted in Figure 7,would curate digital models and the supporting data-bases. It would contain ‘‘user manuals’’ for the digitalmodels so that less experienced users could quicklybegin to use models of interest. It would serve asgatekeeper for those authorized to access the digitalmodels and databases in line with a defined governancemodel, and employ the latest cyber security practices.[18]

Fig. 6—René 65 development timeline.

Fig. 7—Cyber collaborative infrastructure supporting materialsdevelopment.


It would ease collaboration with those outside acompany’s proprietary walls, such as academia, suppli-ers, and partner OEMs. These access agreements wouldidentify specific tools and databases that an authenti-cated user would be able to access. Such an ecosystemwould also ensure that developed models and databaseswould be in a centralized repository rather than scat-tered in multiple locations, and would insure that usersalways had access to the latest version of the model andits associated database.

VI. CONCLUSIONS

Over the ages, the materials community has con-tributed immensely to mankind’s well-being. In thismodern era, the expectation is that material advance-ments will play a major role in fulfilling societal needs ata faster pace. Rapidly improving digital computationaltools and computer technology are making it possible toestimate performance of new materials in applications.Process models are meaningfully reducing the iterationsrequired to achieve full-scale production as well asrapidly assessing the capability of new process tech-nologies. However, the key to incorporation of materialsinto applications is satisfying enterprise needs, not justproviding a superior materials solution.

Accelerated materials development programs canreduce the time for development by concurrency oftasks within the three technology development phases(Materials Feasibility, Materials Demonstration, andMaterials Maturation) rather than accomplishing eachphase faster. All critical development steps are com-pleted. This scenario entails prudent assessment of therisks of going forward and planning for alternateapproaches, especially since every development programinevitably experiences some surprises as the effortproceeds.

Faster development and product insertion of newadvanced materials requires a more comprehensivestrategy than digital computational tools, althoughthese tools are a necessary analysis adjunct. Even ifthe digital tools were perfectly accurate, faster insertionis not a foregone conclusion. An enterprise perspective isstrategically important as many factors influence theviability of a new material in a given application.Success with inserting materials indicates that thesethree strategic elements are necessary in addition todigital tools:

1. Forward-looking materials strategy—a feasible menuof material and process advancements that couldsignificantly benefit a specific application. It isestablished by the inventive genius of materials ex-perts, together with other experts, to explore andselect new materials and process technology based onapplication value.

2. Robust management of materials development—aflexible, disciplined process for maturing and insert-ing new technology into a product in which thematerials technology meets an identified need with anattractive value case.

3. Formation of small, focused high-performance teamsto execute the development programs, composed ofpeople skilled in innovation with the expertise neededto insert the new technology into an application. Theteam members are drawn from materials engineering,design engineering, manufacturing, systems engi-neering, etc.—people are the most critical ingredientin successfully conducting accelerated developmentprograms.

Recent accomplishments with accelerated materialsdevelopments, such as developing a new single crystalturbine blade alloy, René N515 in 2 years and the newturbine disk alloy René 65 in 4 years at unprecedentedspeeds are testament to the viability of prudent accel-erated materials development and demonstrate thepotential of what is possible, realizing the long-standingdream of the materials community to respond quicklyand expertly to satisfy critical product needs. Theaccelerated programs were undertaken to meet a criticalbusiness need by a certain insertion date; the timelinecould not be achieved by following the conventionalsequential development cycle. Also, success with insert-ing materials following the conventional timeline furtherreinforces the viability of applying digital tools inconcert with the additional three strategic elements.

ACKNOWLEDGMENTS

This work was primarily funded by internal GEAviation funds. The long-standing support of DavidL. Joyce, President and Chief Executive Officer, wasessential to success and is gratefully acknowledged.Also, crucial was support from the Office of NavalResearch (ONR), the Air Force Research Laboratory(AFRL), and the Defense Advanced Research ProjectAgency (DARPA) for programs that explored ad-vanced ideas. Special recognition is due to John Politesand Mark Pearson of GE Aviation Engineering whoenthusiastically supported and funded much of thiswork as they oversaw advanced engine developments,and to Steven Wax, who while at DARPA stronglyencouraged the involvement of design engineers inDARPA’s Accelerated Insertion of Materials program.And importantly, this work would not have been pos-sible without the creative energy and dedication on thepart of the men and women of GE Aviation’s Materi-als and Process Engineering Department and GE Cor-porate’s Global Research Laboratory.

REFERENCES1. H. Brooks: Metall. Trans., 1972, vol. 3, pp. 759–68.2. White House Office of Science and Technology Policy: Materials

Genome Initiative for Global Competitiveness White Paper, June2011.

3. J. Allison, D. Backman, and L. Christodoulou: JOM, 2006, vol. 58(11), pp. 25–27.

4. National Research Council: Integrated Computational MaterialsEngineering: A Transformational Discipline for Improved Compet-


itiveness and National Security, National Academy Press, Wash-ington, 2008.

5. National Research Council: Accelerating Technology Transition:Bridging the Valley of Death for Materials and Processes in theDefense Systems. The National Academies Press, Washington,2004, p. 16.

6. D.J. DeTurris and S.J. D’Urso: AIAA Aerosp Am, 2015, vol. 53(10), pp. 32–36.

7. T.J. Kelly, M.J. Weimer: GE Aviation, Cincinnati, personalCommunication, 2008.

8. Technology Readiness Assessment (TRA) Guidance, Prepared bythe Assistant Secretary of Defense for Research and Engineering(ASD(R&E)), April 2011.

9. Manufacturing Readiness Level (MRL) Deskbook version 2.0,Prepared by the Office of the Secretary of Defense ManufacturingTechnology Program in collaboration with The Joint Service/In-dustry MRL Working Group, May 2011.

10. National Research Council: Accelerating Technology Transition:Bridging the Valley of Death for Materials and Processes in the

Defense Systems. The National Academies Press, Washington,2004, pp. 1–2.

11. J.R. Katzenbach and D.K. Smith: The Wisdom of Teams, HarvardBusiness School Press, Boston, 1992.

12. V.J. Russo and D.J. Black: The Splendid Leader, BookFactory,Dayton, 2008.

13. P.J. Fink, J.L. Miller, and D.G. Konitzer: JOM, 2010, vol. 62 (1),pp. 55–57.

14. D. Konitzer: GE Aviation, Cincinnati, personal communication,June 2014.

15. D. Konitzer: GE Aviation, Cincinnati, personal communication,Jan 7 2016.

16. M. Lasonde, J. Heaney, A. Powell, C. O’Brien, B. Bond: 8thInternational Symposium on Superalloy 718 and Derivatives,Wiley, Hoboken, pp. 67–77.

17. D .Mourer, K. Bain, M. Lasonde: GE Aviation, Cincinnati, per-sonal communication, June 2014.

18. R. Irving: GE Global Research, Niskayuna, personal communi-cation, Apr 2012.


Outline placeholderSec1Sec2Sec3Sec4Sec5Sec6

Sec7Sec8Sec9

Sec10Sec11AcknowledgmentsAcknowledgments

materials for a non-steady-state world...2014 distinguished lectureship in materials and society...

Documents