Powder metallurgy of titanium alloys

Powder -metallurgy of titanium alloys F. H. Froes and D. Eylon The powder metallurgy of titanium alloys is reviewed for both the blended elemental (BE) and prealloyed (PA) approaches. The BE technique allows low cost processing with mechanical behaviour at ingot metallurgy levels except for fatigue performance; and this latter property can be raised significantly by use of low chloride sponge. The PA method leads to mechanical behaviour in all respects at least at ingot metallurgy levels, and could soon be accepted as a cost effective manufacturing technique in the demanding aerospace industry. Future developments in the titanium powder metallurgy arena are proposed for both major approaches. IMR/209 © 1990 The Institute of Metals and ASM INTERNATIONAL. Dr Froes is in the Institute for Mater- ials and Advanced Processes, University of Idaho, Moscow, 10, USA, and Professor Eylon is in Graduate Materials Engineering, School of Engineering, University of Dayton, Dayton, OH, USA. Introduction Powder metallurgy (PM) is the production, pro- cessing, and consolidation of fine metal particles into usable engineering components. For many years a large, well established PM industry invol- ving materials such as iron and copper has been in existence. 1,2 The main driving force here is cost reduction while mechanical properties play a secondary role. During the past 15 years the PM of titanium alloys has been established. Cost and material savings have been the major goals3-17 using two approaches - the pre alloyed (PA) tech- nique and the blended elemental (BE) method, based mainly on Ti-6AI-4V (Refs. 18-42), the 'work-horse' alloy of the titanium aerospace industry. These recent advances in titanium PM technology have been associated with alloyed titanium rather than commercially pure material. The PA approach is generally used to produce demanding aerospace components in which mechanical property levels, particularly fatigue behaviour, equivalent to cast and wrou§ht ingot metallurgy (1M) levels are needed.8-1 ,18 This requirement appears to have been met with the establishment of commercial clean powder produc- tion and handling procedures.7-10,16 In contrast, most titanium compacts produced to date by the BE method have not yet achieved the fatigue strength levels required for critical aerospace components because of inherent salt content and/or porosity7-10,18 although recently considerable pro- gress has been made.19-24,31However, production cost for this technique is lower than for the PA technique which has helped develop commercial applications. In the past few years, it has been es~ablished that rapid solidification (RS) offers. some of the same advantages for titanium systems as those already demonstrated for other systems such as alumin- ium,7 In particular, RS allows extended .solid solubility, formation of metastable phases, structure refinement, and freedom from segregation/work- ability problems.7,43-50 The present paper does not cover the RS aspects of titanium powder metallurgy which will be the subject of a separate article.49 This review will concentrate on the BE and PA methods as applied to conventional titanium alloys, with emphasis on the Ti-6AI-4V composition. Other processing techniques such as mechanical alloying51-56 and injection moulding57 of titanium powders, now in the early research stages of development, are also not covered in this review. Blended elemental Powder production The BE approach is basically a conventional press and sinter method.8-10,18,31 Since titanium alloy components are often intended for fracture and fatigue resistant high performance -applications, efforts have been made in the past few years to produce compacts with close to 100% density for enhanced behaviour. The emphasis has been to develop processes resulting in reduced salt content and/or porosity. Most of the BE products are currently made from a mixture of titanium 'sponge fines' (Fig. 1) and elemental or master alloy powder. 8-10,18,23,25, 29,31,58-63The sponge fines are the small (typically -100 mesh (-150 !lm)) irregular shaped particles of nominally pure titanium which are produced during the conversion of titanium ore into an ingot. The commercial production of titanium metal involves the chlorination of natural or synthetical~ produced rutile (Ti02) in the presence of carbon. 2 The most important reaction is Ti02(s) + 2Ch(g) + 2C(s)~ TiCI4(g) + 2CO(g) J'he resulting tetrachloride is then reduced to metallic titanium by the Kroll magnesium process63 or the Hunter sodium process.58 The basic reactions are as follows 2Mg(s) + TiCI4(1)~ Ti(s) + 2MgCh(s) (Kroll process) 4Na(1) + TiCI4(1)~ Ti(s) + 4NaCI(s) (Hunter process) Most of the remnant chloride is removed by vacuum distillation or, after cooling, by water leaching. The final sponge fines typically contain 0·12-0·15 wt_%CI. A typical screen analysis and 162 International Materials Reviews 1990 Vol. 35 No.3 Froes and Eylon Powder metallurgy of Ti alloys 163 I I a sponge fines produced by Hunter process (Ref. 58); b crushed hydrided ingot material; c AITi process sponge (Ref. 59) 1 Titanium powders used in BE method . Weight Element percent AI Mesh size (US) Particle size, Ilm retained V +80 +177 0 0 -80 +100 -177 +149 0·1 N -100 +140 -149 +105 11·2 H -140 +200 -105 +74 32·9 C -200 +230 -74 +64 5·0 Fe -230- +325 -64 +45 23·3 Na -325 -45 17·5 CI chemical analysis of sponge fines produced by the Hunter process are given in Tables 1 and 2, respectively. Several years ago, sponge fines with less than 150 ppm chloride were Eroduced by an electrolytic process in a pilot plant 2,64and recently there has been a renewed interest in this approach to sponge production using a more optimised process.65,66 Commercially pure (CP) titanium powders with chloride levels less than 10 ppm can now be obtained either by crushing hydrogenated in}ot material (Fig. 1b) followed by dehydrogenation6 or by the AITi process which is based on the reaction of Ti02 with fluorine salts followed by reduction with aluminium in a zinc carrier medium (Fig. 1c). The change in morphology of these titanium pow- ders (Fig. 1) result in variations in flow, tap density, and compactibility behaviour. To date the sponge fine powder is used most commonly because of its lower cost and commercial availability. Alloys investigated To date most studies and development work have involved the Ti-6AI-4V alloy7,i8,23,28,29,31,34, 60,61,68-78in a similar way to the titanium net shape casting industry. 79-81 However, work has been reported which indicates that the technique is equally applicable to 1M alloys such as Ti-6AI-6V- 2Sn (Refs. 74, 82), Ti-6AI-2Sn-4Zr-2Mo (Ref. 83), Ti-6AI-2Sn-4Zr-6Mo (Refs. 74, 82), Ti-10V- 2Fe-3AI (Refs. 82, 84-86), Ti-4·5AI-5Mo-1·5Cr (CORONA 5),~' and Ti-5AI-2Cr-1Fe (Ref. 87). Table 1 Screen analysis of sodium reduced Ti sponge fines Recently, ordered alloys based on ThAI + Nb have also been successfully produced by the BE method.88,89 However, to date, no alloy develop- ment efforts have attempted to utilise the specific characteristics of the BE process. Powder consolidation After blending to the desired bulk alloy compos- ition, the powder is cold compacted at pressures up to 400 MPa by either a mechanical press with hard tooling or cold isostatic press (CIP2 with soft tooling to 85-90% 'green' density.31,60, 1 Relatively small plan form (i.e. no re-entrant angles) components can easily be produced by mechanical pressing, which is a high volume, low cost method. Larger and more complex shaped products need to be pressed by the CIP process.21,31 The green compacts are vacuum sintered at about 1260°C to a density of 95-99%. Careful control of powder particle size and size distribution can lead to 99% sintered density without a need for a secondary pressing operation (Fig. 2a) (Ref. 68). A further increase in density can be achieved by hot isostatic pressing (hipping) after sintering, leading to relative densities as high as 99·8% with a standard sponge fine blend. This additional HIP results in improved property levels.18,28,29,31,61Since the BE sintered compacts have relatively low levels of porosity, the parts are hipped without containment in a similar Table 2 Typical chemical analysis (wt_%) of Hunter process Ti sponge fines and sinter com- pacted Ti-6AI-4V produced from these fines and AI-V master alloy powder Sponge Sintered fines compact 6·2 4·1 0·13 0·24 0·03 0·016 0·07 0·002 0·02 0·02 0.02 0.18 0·10 0·10 0·13 0·12 International Materials Reviews 1990 Vol. 35 No.3 164 Froes and Eylon Powder metallurgy of Ti alloys a mechanically pressed and sintered (99% dense); b isostatically pressed and sintered (99·8% dense); c extra low chloride fully dense hipped compact; d thermochemically treated; e treated to produce broken up structure 2 Microstructures of BE Ti-6AI-4V compacts· manner to HIP densification of titanium cast- ings.79-81 The HIP operation increases the density with typically about a 10°/0 increase to the process cost. A 99.8°10 dense CIP + HIP compact micro- structure is shown in Fig. 2b. The chloride in the BE material prevents achievement of 100°/0 density after secondary pressing69,70,76even with chloride levels as low as 0·016 wt_Olo.23,7G-72 Post-sinter densification can also be achieved using a 'soft' tooling method (Ceracon process) utilising a granular pressing medium around a sintered compact and conventional hot forging equipment. 90 Recently, a chloride free «0·0010 wt-Olo) powder produced by the hydride-dehydride (HDH) processb7 has become available and 100°/0 density can now be achieved in CIP + HIp· BE material (Fig. 2C).19-22 A new approach to enhance the density was recently demonstrated by adding a beta quenching stage between the powder sintering operation and the compact hipping. A density of close to 100°10 was reported in compacts containing as much as 300 ppm chloride.87 International Materials Reviews 1990 Vol. 35 No.3 Product forms Near net shapes (NNS) Very complex shapes, such as the impeller shown in Fig. 3a or the airframe component in Fig. 3e can be produced by cipping using elastomeric moulds. However, part size is presently limited to a 60 cm diameter by the availability of cipping equipment. Dimensional tolerances are about ± 0·5 mm. The press consolidation technique employing rigid tools is capable of producing parts up to 1·3 m2, but the shape making capability is probably not as flexible as in the CIP method. The Ti-6AI-4V mirror hub shown in Fig. 3c is an example of a part made in rigid tools by the press consolidation method. Larger BE components are difficult to fabricate by welding smaller parts because this material does not weld well owing to the inherent salt content and/or porosity. 18 However, recent work has shown that weldability can be improved when the chloride level is below 150 ppm (Ref. 91) and the recently developed BE titanium powder compacts with less Froes and Eylon Powder metallurgy of Ti alloys 165 ANr£AL• SPONGE FINES I MASTER ALLOY \ DIRECT POWDER ROLLING f FINAL FOIL PRODUCT---_.~ 4 Schematic for producing rolled titanium alloy foil from BEpowder (Ref.93) r Blended elemental forging preforms offer: (a) inexpensive forging stock, and (b) production of a more equiaxed fatigue resistant microstructure when compared with unworked net shape BE products.97-101Compacts isothermally forged up to 80%, using both regular and low chloride (elec- trolytic) sponge, exhibited a microstructure with a significantly reduced alpha aspect ratio, but because of residual porosity limited fatigue life enhance,-' ment was recorded.70 Compact microstructures Net shapes Typical Ti-6AI-4V BE net shape compact micro- structures produced by mechanical pressing60,73and isostatic compaction23,61 exhibiting residual porosity are shown in Fig. 2a and b, respectively. This residual ~orosity cannot, be healed even by hot working. 9,70 The mechanically pressed and sintered compacts, produced by, one commonly used process 75 display a low aspect ratio alpha structure (Fig. 2a). This is a consequence of the very small beta grain size resulting from the particular process used and the grain boundary pinning effect of the residual porosity. 102,103The alpha plate structure of the isostatically compacted material61 (Fig. 2b) is more typical of material sintered above the beta transus, where a large beta grain size develops, and subsequently cooled slowly to form an alpha plate colony structure. The recently introduced 100% dense BE material displays even coarser beta grain size and alpha plate structure (Fig. 2c), because of the lack of a pinning effect of the porosity during sintering.19 Microstructural refinement The basic microstructures of BE compacts can be modified along with a concurrent improvement in tensile and fatigue strengths by use of an innovative heat treatment (long time anneal low in the ~lpha + beta phase field) to produce a refined broken up structureI9,22,23,104shown in Fig. 2e. Another method for refining the microstructure is temporary alloying with hydrogenI9,22,105-112(Fig. 2d). This thermochemical processing (TCP) method a an impeller (courtesy of Dynamet Technology); b production run BE Ti-6AI-4V missile housing preform and finished product' (courtesy of Dynamet Technology); c production run BE Ti-6AI-4V net shape 35 mm mirror hub (courtesy of Valform); d production run BE lens housing blank and finished product (courtesy of Clevite Industries); e prototype BE Ti-6AI-4V 100% dense airframe component produced from CIP + HIP HDH powder (courtesy of Dynamet Technology); f Ti-6AI-4V pivot fitting for McDonnell-Douglas F"":18 fighter plane (courtesy of MPIF) 3 Components produced by BEmethod than 10 ppm chloridel9-21 should be weldable just like 1M material of similar composition. Direct processing Powder 'metallurgy component production generally involves powder production and consolidation into a NNS as separate steps.7 However, cost reduction is possible by converting the powder directly to a mill product. 8-11,18,90,92-96This has been done using the BE approach to produce foil, sheet, and plate 71,72The plate was produced by rolling a compacted billet; the foil and sheet were fabricated directly from powder as shown schematically in Fig. 4. Forging preforms Another approach to NNS is to use powder compacts as forging preforms in association with NNS forging. In complex shaped aerospace component forgings, the preform represents a substantial portion of .the final product cost. International Materials Reviews 1990 Vol. 35 No.3 166 Froes and Eylon Powder metallurgy of Ti alloys N 900 IE z 2: I 800 I- Froes and Eylon Powder metallurgy of Ti alloys 167 Table 4 Cost per kg for elemental Ti powders* and BE Ti-6AI-4V componentst * Based on a survey of four component producers. t Based on components -0·5 kg in weight. :j:Additional cost over conventional compact because of higher cost of low chloride powder. produced in 10 000 part runs by two manufacturers. The mirror hub shown in Fig. 3c is another example of a currently produced aerospace component which showed a 60% saving over 1M parts. The net shape part shown in Fig. 3f is a fighter plane (F-18) airframe structural component (pivot fitting) made of Ti-6AI-4V which won the 1987 MPIF Excellence Award for demonstrated substantial cost savings. This part has 97% density meeting the minimum tensile strength requirements for this application. Two additional aerospace BE components, which are shown in Fig. 3d and e, are' a lens housing and a prototype airframe component.17 An assortment of commercially pure titanium components for corro- sion resistant applications is shown in Fig. 8a. Low density (20-80%) titanium filters for electrochemi- cal and other corrosion resistant applications are shown in Fig. 8b. Recently, consideration of BE parts for automobile applications such as a retainer cap and a connector rod has indicated that substan- tial weight savings could result (70% for the retainer cap) .117 Penetration of the very high volume automobile market could greatly increase the sales base for BE products and thereby reduce cost in other industries such as aerospace and chemical processing. 45-50 60-100 70-120 30-90 Cost range, US$ kg-1 10-20 120-160 Product Sponge fines (0·15 wt-% CI) HDH CP titanium powder «0·001 wt-% CI) Mechanically pressed compact (no HIP) Cold isostatically pressed (CIP) compact (no HIP) CIP + HIP compact / 100% dense compact (HDH powder), (CIP + HIP):j: 2: ::J 2: 400 X« 2: 200 N 'E z 2: 800 - 1000 tf) tf) w g: 600 If) equivalent to 1M material19,21 (Figs. 6 and 7). This improvement has made possible the use of BE components in critical aerospace applications. The fatigue strength of both 99·8 and 100% dense materials can be improved by innovative heat treatments such as the BUS treatment (Fig.2e) and' thermochemical processing (Fig. 2d) described earlier,19,108 and by the technique in which the material is quenched from the. beta phase field before hipping.87 The less than 100% dense material is lower in cost than the fully dense material and should therefore find use in appli- cations· where cost, rather than mechanical properties, is the major concern. Other mechanical properties such as fracture toughness 71,72,83 and fatigue crack growth rate69,71,72appear to be at the same levels as expected from 1M material of the same chemistry and microstructure when the density is at least 98% of theoretical. 105 106 107 CYCLES TO FAILURES (Nf) 7 Fatigue data scatterbands of conventional BE,. low chloride BE, treated low chloride BE, and PA, compared with wrought annealed material Weldability One problem with BE PM parts, which may limit the range of applications, is the difficulty in welding. The chlorides, as low as 100 ppm, and associated porosity of BE PM compacts cause a severe sputtering during fusion welding leading to a highly porous weld zone. As a result BE PM parts cannot be used for applications requiring a built up structure by welding. Economics and applications The most attractive aspect of the BE technology is its relatively low cost. Approximate cost figures are given in Table 4, which is based on a survey among several titanium BE component producers. Because of the relatively low cost, several BE components are in use, both in aerospace and in commercial markets. The component shown in Fig. 3b is a BE Ti-6AI-4V missile housing which was Prealloyed Powder production In the PA approach the starting stock for powder production contains the required alloy elements. Three general methods of powder production are presently available: (a) comminution; (b) melting, followed by atomisation, and subsequent cooling; and (c) coreduction. Other emerging processes which relate more to RS technology will be discuss- ed elsewhere.49 In the comminution process bulk titanium alloy or alloy turnings are embrittled by hydrogenation, crushed, and subsequently dehydrogenated. This process, which has been discussed in the 'Blended elemental' section above, is termed the hydride- dehydride (HD H) process. 118 The powder is angular (Fig. 1b), typically with 700-800 ppm increased oxygen content over starting stock. International Materials Reviews 1990 Vol. 35 No.3 168 Froes and Eylon Powder metallurgy of Ti alloys 8 a assortment of BE commercially pure titanium parts for chemical industry and b assortment of porous BE titanium filters for electrochemical processes (courtesy of Clevite Industries) Unless a great deal of care is exercised production practices can introduce contaminants; however, the relatively low cost «$11.00 kg-1 for hydrogenated powder in large quantities119) is attractive. For the production of high integrity NNS, clean spherical PA powder is generally required.8-1o,18 However, because of the extreme reactivity of the molten metal, and high melting point, titanium alloy powders free of contamination cannot easily be produced by the atomisation processes routinely practiced for less reactive metals with significantly lower melting points, such as aluminium.7,120 This has led to the development of a number of atomisation processes in which only local melting occurs. These processes include the rotating elec.- trode process (REP), electron beam rotating disc (EBRD), powder under vacuum (pulverisation sous vide (PSV)), continuous shot casting (CSC), rotat- ing electrode atomisation (REA), the Colt- Crucible hydrogenation (C- T) process, the gas atomisation (GA) process, and the pendant drop (P-D) methods.~10,18,28,57,120-123All but the last two processes involve spherical powder production by centrifugal atomisation. The C- T process yields a spherical powder while the P-D method generally produces elongated particles.122,123Only the REP process is at a full production status, using a plasma heat source (PREP) to yield powder (Fig.' 9a) that is not contaminated with tungsten.124 The REA process57 and the GA process in combination with clean room facilities, are about to reach production status. The PSV process appears promising but can suffer from the presence of aluminium rich shells on smaller particles, which can degrade fatigue behaviour. 125,126 Because of the relatively high cost of titanium powder, attention has been given in recent years to the production of prealloyed powder directly from chemical precursors such as Ti02 or TiCl4 which are relatively cheap starting materials.62 It appears that there are definite prospects usin~ I?roces~es such as the Hurd Shaker process127-1 9 In WhIch titanium, vanadium, and aluminium chlorides are reacted to produce an alloy powder (Fig~ lOa). Also a as produced; b strain energised (SEP) 9 PREP Ti-6AI-4V powder International Materials Reviews 1990 Vol. 35 NO.3 ,10 a Ti-6AI-4V powder particle produced by Hurd Shaker process (Refs. 127, 128) and b powder particles produced by Goldschmidt method (Refs. 130, 131) redu~tion of mixed oxides using calcium (Gold- schmIdt method)130,131 (Fig. lOb) plasma tech-. 132-135 ~ . , .niques, molten salts, 1 6 molten fluonde salt processing, 137,138electrolytic Rrocessing,64-66,139-141 or the AITi-oxy process,59 may be viable approaches to the production of pre alloyed pow- ders. Development of a low cost, high quality direct powder production technique could significantly accelerate the commercialisation of the PM tech- nology, particularly for aerospace applications.142 Powder cleaning Unlike other common metal powders (such as steels), titanium powder cleanness is based on ultra-clean powder making processes.124 A number of powder cleaning processes, directed towards removing discrete contaminant particles, have been evaluated for titanium alloys in the past. However, these have not proved to be successful, and consequentll they have not seen commercial imple- mentation.1 ,21It is now well established that clean powder production is- strongly preferred to the cleaning of contaminated powder. Alloys investigated In addition to extensive studies on Ti-6AI-4 V (Refs. 7-10, 18, 28, 29), work has also been carried Froes and Eylon Powder metallurgy of Ti alloys 169 out on Ti-6AI-6V-2Sn (Refs. 143-147), Ti-6AI- 2Sn-4Zr-6Mo (Refs. 148, 149), CORONA 5 (Ti- 4'5AI-5Mo-1'5Cr) (Refs. 150-152), Beta III (Ti- 11·5Mo-6Zr-4·5Sn) (Ref. 153), Ti-10V-2Fe-3AI (Refs. 85, 86, 154), the high strength Ti-185 (Ti-1AI-8V-5Fe) alloy (Refs. 155, 156), the high temperature alloys IMI-685 (Ti-6AI-5Zr-0·5Mo- 0·25Si) (Refs. 157-159) and IMI-829 (Ti-5·5AI- 3·5Sn-3Zr-1Nb-o·3Si) (Ref. 160), and the intermetallics Ti3AI and TiAI (Refs. 161-168). In all cases, the use of clean powder allows mechanical properties equivalent to 1M levels to be achieved. Recent work has shown that using the PREP pr'ocess it is possible to produce a segregation free beta alloy Ti-1AI-8V-5Fe (Ti-185) (Refs. 155, 156, 169) with a very high tensile strength (UTS 1482 MN m-2, elongation 8%). The production of this alloy is almost impossible using 1M methods bec~use of significant iron segregation which is detnmental to mechanical properties. The addition of high~r lev~ls of Si (1!P to 0·4 wt_%) to near alpha alloys lIke TI-6AI-2SI-4Zr-2Mo for an. improved creep strength also became possible by using the' PREP process. 170 Powder consolidation Powder consolidation can be achieved by a number of processes including hipping,8-10,18,171fluid die compacting (FDC),8-10,18 the Ceracon process,90 vacuum hot pressing (VHP), 106and extrusion.156 In the HIP process, pressure and temperature are applied s.imultaneousl.y inside an autoclave allowing full denSIty to be attaIned171 as shown in Fig. 11. In the FDC process18 shaped cavities or dies behave as a .viscous liqu.id under pressure at temperature (~Ig. 12). In thIS case the pressure can be applied in eIther an autoclave or forge press. The Ceracon proc~ss make~ use of soft tooling using a granular pressIng medIum and conventional hot pressing. The VHP method utilises hot compaction of pow- der in a forge press,106,172adapted to a vacuum system, using permanent and reusable dies to produce fully dense parts. Compaction of titanium powders by extrusion is carried out with either loose powder or precompacted powder inside a can. here, hot processing produces a fully dense pro- dUCt.156,170An advantage of this last process is that a lower temperature can be used for compaction, and there is more relative motion between the particles during compaction than in the other four methods mentioned above.173 High pressure hipping Recently HIP equipment with high pressure capabilities has become commercially available with some units able to achieve pressures up to 300 MPa. Work done on PREP Ti-6AI-4V and Ti-10V-2Fe-3AI (Refs. 174-177), demonstrated that at higher pressures full compaction and. bond- ing can be accomplished at temperatures as low as 650°C. The low compaction temperature develops a very fine microstructure which exhibits excellent combinations of tensile strength and elongation for the Ti-6AI-4V alloy (Table 5)._ International Materials Reviews 1990 Vol. 35 NO.3 170 Froes and Eylon Powder metallurgy of Ti alloys 11 Schematic of Crucible ceramic mould (CCM) process (Ref. 18) Rapid omnidirectional compaction (ROC) Rapid omnidirectional compaction (ROC),178 a modification of the FDC process,8-10,18is emerging as a viable alternative for compacting pre alloyed titanium powder into full density complex shape products with a range of microstructures that provide good combinations of mechanical proper- ties. Rapid omnidirectional compaction is a compaction process in which the powder is sub- jected to a pressure close "to isostatic which is developed in a conventional hot" forging press.178 The method utilises a· thick walled powder con- tainer called a 'fluid die' which is made from an alloy capable of plastic deformation at consolid- ation temperature and pressure. These are typically 900-950°C at 850 MPa for Ti-6AI-4V (Refs. 179, 12 Schematic of fluid die powder compaction (FOC) process (Ref. 18) International Materials Reviews 1990 Vol. 35 No.3 180) and Ti-10V-2Fe-3Al (Ref. 181) alloys. Recently, it was demonstrated that when a dwell cycle of 15 min at the bottom of the press stroke is added, a full density compaction can be achieved at temperatures as low as 650°C (Ref. 182). This modified process is termed IsoROC. As a result of the very low compaction temperature diffusion controlled reactions are minimised; for example, precipitate coarsening, grain growth, or reaction between dissimilar materials in hybrid concepts. This in turn leads to improvements in prop- erties21,179-182similar to those observed in HIP compacts produced using high pressure, low tem- perature HIP compaction cycles.174-177 Product forms Near net shapes Production of complex NNS is possible using one of five competing techniques: metal can, ceramic mould, 18-~7 fluid die,8-1o,18 electroplated nickel can,183 or the Ceracon process. 90The metal can is shaped by state of the art sheet metal forming , methods such as brake bending, pressing forming, spinning, or superplastic forming .. The ceramic mould process relies basically on the technology developed by the investment casting industry79,80in that moulds are prepared by the lost wax process (Fig. 11)8-10,18,18By combining this process and PM hipping, low cost complex configuration NNS can be produced.15,16,27,184 A typical component produced by the ceramic mould process is shown in Fig. 13. The FDC process18 involves production of shaped cavities or dies which are then filled with powder. In the nickel can process, nickel is elec- troplated on a wax core which is subsequently Froes and Eylon Powder metallurgy of Ti alloys 171 Table 5 Tensile properties of Ti-6AI-4V and Ti-10V-2Fe-3AI PREPpowders* (Refs. 174, 175, 177) Compaction Heat treatment, t 0·2% YS, UTS, Alloy temp.,oC °C/h MN m-2 MN m-2 EI.,% RA,% Ti-6AI-4V 650 As compacted 1082 1130 8 19 Ti-6AI-4V 650 815/24/AC 937 1013 22 38 Ti-10V-2Fe-3AI 600 As compacted 951 992 14 49 760/11WQ + 510/8/AC 1226 1295 3 6 * Compacted by HIP under 300 MPa pressure for 24 h. t AC air cooled; WO water quenched. removed by heating and the cavity so produced is used to make the required shape.183 The Ceracon process allows essentially isostatic pressing to be a~- plied tq complex shapes yielding fully dense parts. 0 Direct processing Processing of titanium pre alloyed powders into mill products can be achieved and Ti-6AI-4V plates have been produced.42,71,72,185 The specific method used was to fabricate a PM flat preform by HIP compaction, which was then rolled to final plate dimensions. The mechanical properties produced were very attractive.71,72 Since the PM preform can be hipped into a slab shaped compact, the rolling reduction required for such a product will be minimal, with a potential for cost savings. a c b d a F-14 fuselage brace produced by CCM process; b F-15 keel splice former (CCM); c F-18 engine mount support fitting (CCM); d helicopter fitting and an airbus connector arm produced by MMB 13 Ti-6AI-4V PA components In~ernational Materials Reviews 1990 Vol. 35 NO.3 172 Froes and Eylon Powder metallurgy of Ti alloys Forging preforms Production cost analysis of isothermally forged components indicates that a large portion of this cost is involved in fabrication of the preform, which is then isothermally forged.186 Consequently, the use of Ti-6AI-4V (ReL 182) and Ti-l QV-2Fe-3 Al (Refs. 84-86) powder compacts as preforms was investigated recently. Results to date indicate that this appears to be a viable approach to production of high integrity, low cost isothermal forgings with properties equivalent to or exceeding those of wrought products.8-10,28,29 The combination of powder compaction and forging provided a greater latitude for microstructural modification35,84--:86 when compared with products froduced by a true net shape PM approach. 8-:-10,28,2 Compact microstructures Net shapes Cost considerations strongly favour production of net shapes or NNS in complex shaped titanium alloy components.8-10,18 Use of hipping, VHP, FDC (Fig. 11), or other NNS techniques generally preclude significant working of the powder during consolidation. Net shape processing is incompatible with cold or hot working of the product after consolidation. Thus', conventional methods of con- trolling microstructure in titanium alloys are not available. Hipping of Ti-6AI-4V below the beta transus temperature results in a microstructure consisting of alpha plates in a beta matrix.8-10 The aspect ratio of these plates is predominantly determined by the amount of work that the alpha (or precursor martensite) receives,187-191and this is not large in the case of hipping (Fig. 14a). It is now well established that fatigue initiation resistance, both at room temperature97-101 and elevated tem- peratures,192-194 is improved with a small equiaxed alpha morphology rather than a coarser/lenticular shape. Modification of the HIP cycle results in a more equiaxed alpha morphology (Fig. 14b). To decrease the alpha aspect ratio further it is neces- sary to increase the amount of strain in the alpha particles to promote relaxation of the alpha.187-191 This can be achieved by deforming the powder before compaction (Fig. 14c) using the strain ener- gising process (SEP),37,149,179-181,195-i97which in- volves deforming the powder particles by rolling (Fig. 9b). The SEPmethod was also found to be effective in promoting formation of equiaxed alpha phase in the Ti-6AI-2Si-4Zr-6Mo alloy. 149Alter- natively a high strain rate compaction method such as FDC, ROC,178-182or compaction of the powder at relatively low temperature and high pressuresl82' can be used to decrease the alpha aspect ratio. After strain energy processing, the as compacted structure (Fig. 14c) shows areas retaining the origi- nal particle shape that were not recrystallised. This is a result of small powder particles that have become attached to larger particles (i.e. satelliting) and then have not been cold rolled because of their smaller size (Fig. 9b). These non-recrystallised areas result in . a marginal fatigue strength International Materials Reviews 1990 Vol. 35 NO.3 improvement of the SEP compacts over REP compacts with the same level of contamin- ants.32,195,196,198However, a fully recrystallised structure was obtained in SEP powder compacted by the FDC process (Fig. 14d) which exhibited good tolerance to contaminants.32 The hot press consolidation method allows attainment of a more uniform recrystallised alpha structure either after VHp172 or by use of the FDC process.18 The microstructures developed during VHP demon- strated the capability to tolerate some porosity without significant loss of fatigue strength.3 ,198Hot isostatic pressing compaction of Ti-6AI-4V PREP powder with different size fractions of particles does not seem to affect the final microstructure or the resulting mechanical properties.199 Microstructural refinement The microstructure of PA compacts can also be modified by the BUS treatment (Fig. 14e)8-10,22,200 discussed above in relation to BE material (Fig. 2e). This BUS treatment (Fig. 14e) can also be .achieved by. compacting the prealloyed powder at lower temperatures in the range of 60o-700°C by ultra-high pressure hipping174-17 or by IsoROC.182 Recently the thermochemicaJ processing (TCP) techniquelO6-110has been used to produce finer and more equiaxed microstructure in PA compacts25 (Fig. 141), which exhibited improved fatigue behav- iour. Hydrogen is temporarily introduced either to the powderr05,106 or to the compactl07-110 up to 2 wt-Olo, and subsequently removed by vacuum annealing to give a fine. alpha structure similar to the BUS treatment results (Fig. 141). This 'micro- structure exhibits superior tensile and fatigue prop- erties in PA Ti-6AI-4V alloy22 and preliminary results suggest that this technique can also be applied' successfully to the intermetallic Ti3AI composition. 156,201 The as compacted microstructure can also be changed by thermomechanical processing,35 in which alpha + beta forging of powder compacts is followed by an alpha + beta solution treatment to improve significantly the low cycle fatigue strength of Ti-6AI-4V (Refs. 35, 202) by changing the predominantly low aspect ratio alpha structures of the as hipped material to a duplex structure of low aspect ratio primary alpha surrounded by a finer alpha in a beta matrix. Thermally induced porosity Another aspect of pre alloyed titanium PM compact microstructure is the thermally induced porosity (TIP). Quite often when compacts are heat treated close to the beta trans us temperature, micrometre size porosity appears.35 This porosity is the result of inert gases like Ar or He which are adsorbed on to the particles during powder production or the HIP can filling. process. A small leak in the HIP container can also result in this type of porosity. This porosity can be easily detected if compacts are heated for a few hours above 1100°C (Ref. 203). These pores are faceted in nature with the pore faces coinciding with low energy planes of the hexagonal alpha titanium (Fig. 15). Froes and Eylon Powder metallurgy of Ti alloys 173 a hipped REP; b CCM hipped PREP; c SEP + hipped; d SEP + FDC; e BUS treated; f TCP treated 14 Microstructures of Ti-6AI-4V Mechanical properties For both 1M and PM titanium alloy products, mechanical properties are strongly dependent on alloy microstructure204· and on foreign particles or contaminants. 7 ,18 These contaminants (if not exceeding certain levels) have little effect on static properties .such as tensile behaviour26 or fatigue crack propagation (Fig. 16),205but can significantly degrade initiation related properties such as fatigue (Fig. 17a).32,34,35,39,40,197,206-210 For 1M material the chance of a foreign particle or contaminant being present is extremely small, but finite. The very small chance of a contaminant related failure in a component is now accepted on a statistical basis. In the PM case the recent improvements in powder making and handling techniques have also led to very low contaminant levels,124 so that the PM approach should be accepted in a similar manner to International Materials Reviews 1990 Vol. 35 No.3 174 Froes and Eylon Powder metallurgy of Ti alloys a macropores; b micropores 15 Hexagonal TIP porosity in PA Ti-6AI-4V compact after 1200°C for 4 h the 1M route. In addition, work is now in progress to improve the NDE techniques for identifying the smallesf gossible inclusions in titanium alloy compacts. 11,212The mechanical properties of low contaminant powder products are now at least equivalent to 1M levels (including welded STRESS INTENSITY (oK). MN m-312 10 102 material,150,151,213see the section 'Weldability' below), most importantly the critical fatigue behav- iour (Table 6 and Fig. 7) .15,26,37,214However, the final hurdle that must be overcome is to convince design engineers that powder products can now be safely used. This can only be achieved by accumu- lation of a substantial amount of. data so that 16 Comparison of FCGR of PM PA and 1M Ti-6AI- 4V (Ref. 205) low contaminate a high and low contaminate; b unseeded and seeded with 50 f-tm contaminates 17 Comparison of fatigue life of Ti-6AI-4V PA N 1000 'E z ~ - 800 V>V> W 0:: I- 600V> ~~ Q.J ~ 400 u u x high contaminate- V> ~ 600 0:: l- V> ~ 400~ 2: X 200 (b) 103 104 105 106 107 CYCLES TO FAILLJRF (N.) , I PREAllOYED I PM CONSOLIOATE~' IAFWAl/METCUTI I FREQ.: 5 Hz 10 oK.ksiJin FCGR OF 1M AND PM Ti-6AI-4V AT ROOM TEMPERATURE' IlABORATORY AIRI R: 0.1 FREQ. : 5 TO 30 Hz Q.J =- u-c: International Materials Reviews 1990 Vol. 35 NO.3 Froes and Eylon Powder metallurgy of Ti alloys 175 Table 6 Properties of Ti-6AI-4V PA powder compacts (Ref. 214) 0·2% YS, MN m-2 UTS, MN m-2 Elongation, % RA, % K1c, MN m-3/2 930 992 15 33 77 statistically significant design curves can be developed. The importance of powder cleanness for good fatigue behaviour is demonstrated in Fig. 17a where compacts produced from clean state of the .art powder are compared with products from earlier less clean powder. 29 Quantitative .information on the influence of contaminant particles on fatigue behaviour was developed in a programme in which very clean PREP powder was 'seeded' with con- taminants of various sizes, shapes, and chemical compositions.36,207-209 It was shown that even 50 ~m contaminants can cause fatigue life degrad- ation (Fig. 17b). This degradation was equivalent in magnitude to fatigue life losses measured in PM compacts with naturally occurring contamin- ates.36,39,40 Mechanical properties can be enhanced using the BUS or TCP processes (Fig. 14e and f).105,106 These processes result in the high cycle fatigue strength of clean Ti-6AI-4V PREP compacts being raised to levels above the best obtained in 1M products (Fig. 7).8-10 A duplex structure of low aspect ratio primary alpha surrounded by a finer alpha in a beta matrix can be produced by alpha + beta forging and solution treatment resulting in a significantly improved low cycle fatigue strength.33,202 Additionally, use of higher strength alloys such as Ti-10V-2Fe-3AI allow tensile strengths and fatigue performance exceeding 1M Ti-6AI-4V material to be achieved.85,86 The static properties of PA PM Ti-10AI-2Fe-3AI were comparable with those of ingot material of the same composition.84-86,181 However, because of the high strength level typical of solution heated and aged metastable beta alloys, a great sensitivity to defects was observed in fatigue testing.84,181Using PREP powder a strength level of 1480 MN m-2 was achieved in the Ti-1· 5AI-8V- 5Fe (Ti-185) alloy with a very acceptable 8% elongation.155,156,169The high fracture toughness titanium alloy CORONA 5 (Ti-4·5AI-5Mo-1·5Cr) produced using rotating electrode powder also exhibited· tensile properties equivalent to those found in ingot products. 152 In this alloy an alpha + beta HIP cycle was found to produce superior property combinations than a beta HIP treatment. An excellent combination of tensile properties was achieved by low temperature HIP compaction which maintained the fine powger particle structure in the compacted article, and hence the combi- nation of high tensile strength and elongation (Table 5) .174,1'15,177 High temperature PM titanium alloys like IMI-829 (Ti-5·5AI-35Sn-3Zr-1Nb-0·3Si) exhibit dynamic and static mechanical properties at room and elevated temperatures at levels equivalent to those of 1M alloys.215 Weldability Detai~ed studies of the' weldability of PA compacts150 ipdicated that there appeared to be no difference between the welding characteristics' and microstructure/mechanical properties developed in welded PM material and 1M products.151,213 Economics and applications Following the strong recommendations made at the Sagamore Conference3 major efforts have been made by the US aerospace industry to produce cost effective, high integrity components using the PA approach. A number of components produced under US Air Force sponsorship are shown in Fig. 13., As previously mentioned, all mechanical prop- erty requirements were met in these parts, and a number of demonstration parts are now flying in the F-15 (keel splice former, Fig. 13b) and F-18 (engine mount support fitting, Fig. 13c). However, to date no bill of materials production part is being produced using the PA PM method. The reasons for' this situation are discussed in the final two sections of this review. Since the cost of titanium component fabrication lies mainly in forging and machining,13,216-218with material cost also a consideration, the selection of appropriate parts can be made only after evaluation of these factors for that specific part. Generally, the PM approach is most attractive for large, complex shape parts with a high buy/fly ratio when fabri- cated by conventional means. Presently, the largest autoclave available (120 cm dia. x 240 cm high,-- working volume) limits the size, unless approaches such as subsequent welding to form larger components are used. Table 7 lists the current forging weight, PM product weight, fin'al part weight, and the anticipated potential cost savings for various parts that have been produced by the PA technique.14,15 These estimates suggest that cost savings with the PA route over forged parts could range between 20 and 500/0depending on the size and complexity of the part and quantity of parts produced; higher volume runs resulting in higher sav- ings.8-10,214,216,217,219 .More recent detailed studes of the economics of the process by Yolton and Mall,220 Witt and Ferreri 154,221Froes and co-workers 17,222Moll 223 Kelto,224 Grewe et al. ,225,226and Winkler154,221:227 have shown that with the correct part choice the PM approach can be significantly less expensive than the 1M method. These studies have shown that this advantage is derived from the enhanced material utilisation over alternative processes such as forged billet, casting, andNNS forging, and stems from the greatly reduced machining required. This conclusion applies to a number of PM NNS International Materials Reviews 1990 Vol. 35 No.3 Forging PM Component 18 Cost savings obtained using prealloyed powder approach compared with conventional forging for complex parts techniques including the ceramic mould and fluid die processes. Grewe and co-workers225,226 used a ceramic die process to produce complex undercut parts such as impellers. Using this technique cost savings up to 40% (Fig. 18) could be obtained for parts such as those shown in Fig. 13d. The conventional forging approach results in the costs given in Table 8. Equivalent figures for the PM route are also given. The obvious ways to reduce the PM part cost even further are to reduce powder cost and to make a closer to net shape component, thereby reducing the machining cost. However, a closer to net shape part may not give the full cost reduction expected from reduction in machining cost since the more complex NNS is likely to be more expensive to inspect. In another analysis of PM part cost, Mo1l223 using the Crucible ceramic mould (CCM) process and the F-18 engine mount support (Fig. 13c), which is a relatively simple shape, has suggested that powder cost and production experience are the major factors in determining cost (Figs. 19a and 19b). At a powder cost of $100 kg-1 the PM part is reduced only marginally below that of the part machined Table 8 Comparison of conventional and powder metallurgy processing costs for impeller* (Refs. 225, 226)' Powder metallurgy US$ % -100 -2·5 400' 11 100 2·5 1670 44 2270 60 Conventional forging Processing step US$ % Forged rodt (19·25 Ib) 460 12 Centre less ground Powder making; Shape makingll Machining and inspection§ 3330 88 Total 3790 100 * Using US$1 = 2·1DM; 1 Ib = 0·45 kg. t US$24 Ib-1, final component weight 3·85 lb. ; -US$100 Ib-1atomisation. II US$2o-40 Ib-\ -US$25 Ib-1 used. Includes qualification and testing. § From German aircraft industry. PM part half cost of forged part. .from a plate even in a full production environment. If powder cost could be reduced to $20 kg-I, Moll's analysis223 indicates that the PM part would cost ----33% that of the conventional part in the same full production mode. It should be recognised that this analysis is for a part which does not meet the degree of complexity required to exploit fully the PM cost advantage. For a more complex part the savings with a $20 kg-1 powder should be even more significant. The analyses discussed above17,222,223,225-227 indi- cate that for large complex parts using ceramic mould-die processes the PM approach can be highly competitive with conventional processing by forging and machining and that the major factors in determining the PM part costs are powder costs, production environment, and nearness to true net shape (reduced machining cost). In fact, a true net shape can be significantly less expensive than a part requiring even minimal machining because of the high cost of machining set up procedures. Similar conclusions were reached by Kelt0224 using the ROC technique. Despite the significant cost savings possible using the PA approach there are virtually no flying applications at the present time, except for a few demonstration parts. The reasons for this situation are discussed further in the final section of this review . 100 : :~;:::::i.... , ~0 80 ··.·tc Hf0-en ........• 0 :::::u 60 ••~r-Cl ~::r:c ......-.: ::: ~t :J :!:::- • ~ • t •u 40- ·t·· .. Metal-cuttingc .....,... : : : :: :J :: ~:~ costc ;;::!c loss of~ 20- mater;al Cost of material (net) 0 International Materials Reviews 1990 Vol. 35 NO.3 Froes and Eylon Powder metallurgy of Ti alloys 177 NUMBER OF PARTS 19a Relative part cost v. number of parts produced at various assumed powder costs for F-18 engine mount support fitting shown in Fig. 13c Current status and future potential The current state of the art of titanium PM has been reviewed in the two major areas of devel- opment: blended elemental and prealloyed. In the BE area low cost production of complex shapes is possible with most mechanical properties equivalent to cast and wrought levels. The excep- tion is fatigue behaviour which is degraded by the salt content/porosity which can be associated with this product. At the same time these features also make this material very difficult to weld. Advances in this area are coming from the use of low chloride sponge or chloride free CP titanium granules in combination with densification operations such as hipping or forging (either conventional or isother- mal) and innovative treatments such as the BUS treatment or thermochemical processing. The BE approach has already been used success- fully to manufacture production runs of com- ponents albeit with fatigue behaviour below 1M levels. The use of low chloride sponge should increase use of this technique especially if the cost can be reduced below the $60 kg-1 currently paid for this input materia1.17 An obvious approach here would be use of the low halide salt AITi-Oxy sponge.59,62 Other advances with this powder method should include development of alloys par- ticularly amenable to this approach. The PA technology is at the point of acceptance as a cost effective manufacturing method by the extremely demanding aerospace industry. 17,219, 228,229 Use of clean powder making and handling methods has allowed mechanical properties at least equivalent to cast and wrought levels to be achieved. However, the PA method is presently more expensive than is desirable, so that devel- opments could include direct alloy powder produc- tion from chemical precursors and lower cost shape making procedures, the latter encompassing computer techniques and the ability to fabricate larger, more complex configurations closer to true· net shape. Advances in NDE techniques could then be included in an overall computer aided design- computer aided manufacture-computer aided inspection (CAD-CAM-CAI) concept of net shape production. Advances should also come from use of higher strength and higher temperature titanium alloys, where control of microstructural features such as grain size should offer advantages over cast and wrought product. Further, the inherent low room temperature ductility of alloys such as the titanium aluminides make forming and machining of complex shapes a very difficult task. Thus a true net shape PA PM approach colild have a great advantage. Additionally, innovative treatments such as the BUS treatment or thermochemical processing should further enhance mechanical properties. Other advances could include use of (a) titanium PA preforms for forging or (b) direct processing; both offer the potential for cost reduc- tion. Graded property components could easily be produced by PM either in combination with cast, or cast and wrought concepts. The slow acceptance of the titanium PA PM technology into the aerospace industry is difficult to understand.17,219,228,229 Here is an approach in which mechanical behaviour, including fatigue, is at least as good as 1M material and parts can be produced at lower cost than by the 1M method. Yet despite these characteristics there is not one pro- duction application. It appears that the ultra- conservative aerospace' industry is extremely reluc- tant to change even if it means getting an equivalent product at a lower cost. For the high integrity cost effective PA approach to become accepted a number of actions must occur. 17 Customers must insist on cost effective products. Design data must be developed at manufacturer, technical society, and military organisation levels. Accompanying this should be design guidelines for use of PM parts. At the same time more technical publicity should be given for the technique and its capabilities. Powder metallurgy parts producers should concentrate on parts that are particularly suitable to this fabrication approach such as bimet- allic parts like integrally bladed rotors or thin complex hollow structures.229 For the former, concepts such as bonding powder to a solid, and International Materials Reviews 1990 Vol. 35 NO.3 full production powder at $50 Ib-1 / ~ full production PM parts pilot production PM parts / 50 40 30 20 10 POWDER COST, $ Ib-1 Relative part cost v. assumed powder cost under various production environments for F-18 engine mount support fitting shown in Fig. 13c development pilot production l- V) o .U I- 0::: ~ W >~« ...J w 0::: I- 0:: ~ l- V>o u w > I- « ...J w 0:: 19b 178 Froes and Eylon Po~der metallurgy of Ti alloys Glossary of terms Acknowledgments The authors would like to acknowledge the contri- butions made to titanium powder .metallurgy science and technology by their many colleagues and co-workers without whom this review would not have been possible. In particular we would like to acknowledge H. Jones, J. H. Moll, and C. F. Yolton. In addition the assIstance of Miss Karen A. Sitzman, Mrs Jan S. Halldorson, and Ms Julia Silver in manuscript preparation is greatly appre- ciated. powder to powder have already been demon- strated. For the latter, a finer grain size than can be obtained in castings is required with far more complexity than can be achieved with forgings, making PM a very attractive approach.229 Also true net shapes, which eliminate machining completely, will further reduce the cost of PM parts compared with conventionally fabricated components. The use of rapidly solidified alloys, such as rare earth or metalloid dispersion strengthened alloys, for high temperature use can only be achieved via PM. This, in combination with lower cost powder making processes, should guarantee use of the PM approach. Added to this could be use of hybrid concepts such as graded compositions or composite reinforced structures which can be much more amenable to PM than 1M fabrication. An inter- esting use of porosity induced by inert gas' has recently been demonstrated.5o,230 By deliberately allowing the inert gas to be present in the compact a porous material with as much as 40% porosity can be produced, this material could have interesting 'space-filling' applications. Injection moulding may also be a viable method for the fabrication of titanium PM parts57 provided a suitable fine powder is available. Finally, PM component use will be accelerated by use in new aerospace systems rather than as replacements in current systems. 1. F. v. LENEL: 'Powder metallurgy'; 1980, Princeton, NJ, Metal Powder Industries Federation. 2. R. M. GERMAN: 'Powder metallurgy science'; 1984, Prince- ton, NJ, Metal Powder Industries Federation. 3. Summary of Air Force/industry manufacturing cost reduc- tion study', US Air Force Technical Report, AFML- TR- LT-73-1, Jan. 1973. 4. J. N. FLECK and L. P. CLARK: SAMPE Q., Oct. 1976, 12, 10. 5. G. I. FRIEDMAN: Int. J. Powder Metall., 1970, 6, (2), 43. 6. G. H. GESSINGER: in 'Titanium '80, Science and technology', Vol. 1, (ed. H. Kimura and O. Izumi), 243; 1980, Warrendale, PA, Metallurgical Society of AIME. 7. F. H. FROES and J. R. PICKENS: J. Met., 1984, 36, (1), 14-28. 8. F. H. FROES and D. EYLON: in 'Titanium, Science and technology', Vol. 1, (ed. G. Lutjering et al.), 267-286; 1985, Oberursel, FRG, Deutsche Gesellschaft fur Metall- kunde. - 9. F. H. FROES and D. EYLON: Powder Metall. Int., 1985, 17, (4), 163-167; (5), 235-238. 10. F. H. FROES and D. EYLON: 'Titanium technology: Present status and future trends', (ed. F. H. Froes et al.), 49-59; 1985, Dayton, OH, Titanium Development Association. 11. F. H. FROES and D. EYLON: in 'Titanium net shape technologies', (ed. F. H. Froes and D. Eylon), 1-20; 1984, Warrendale, PA, Metallurgical Society of AIME. 12. F. H. FROES and D. EYLON: in 'PM aerospace materials', Vol. 1, 39-1 to 39-19; 1984, Shrewsbury, MPR Publishing Services. 13. D. EYLON, M. FIELD, F. H. FROES, and G. E. EICHELMAN: SAMPE Q., 1981, 12, (3), 19-26. 14. D. EYLON, F. H. FROES, and L. D. PARSONS: in 'Proc. 24th structures, structural dynamics, and materials conf.', 586- 593; 1983, New York, American Institute of Aeronautics and Astronautics. 15. D. EYLON, F. H. FROES, and L. D. PARSONS: Met. Powder Rep., 1983, 38, (10), 567-571. 16. D. EYLON, F. H. FROES, L. PARSONS, and E. J. KOSINSKI: Mater. Eng., Feb. 1985,35-37. 17. F. H. FROES, D. EYLON, and R. G. ROWE: in 'Titanium 1986, Products and applications', Vol. 2, 758-781; 1987, Dayton, OH, Titanium Development Association. 18. F. H. FROES, D. EYLON, G. E. EICHELMAN, and H. M. BURTE: J. Met., 1980, 32, (2), 47-54. 19. D. EYLON, R. G. VOGT, and F. H. FROES: in 'Progress in powder metallurgy', Vol. 42, (ed. E. A. Carlson and G. Gaines), 625-634; 1986, Princeton, NJ, Metal Powder Industries Federation. 20. s. ABKOWITZ and D. M. ROWELL: in 'Progress in powder metallurgy', Vol. 42, (ed. E. A. Carlson and G. Gaines), 611; 1986, Princeton, NJ, Metal· Powder Industries Federation. 21. s. ABKOWITZ and D. M. ROWELL: J. Met., 1986, 38, (8), 36. 22. D. EYLON, R. G. VOGT, and F. H. FROES: in 'Modern developments in powder metallurgy', Vol. 16, (ed. E. N. Aqua and C. I. Whitman), 563-575; 1985, Princeton, NJ, Metal Powder Industries Federation. 23. s. ABKOWITZ, G. J. KARDYS, S. FUJISHIRO, F. H. FROES, and D. EYLON: in 'Titanium net shape technologies', (ed. F. H. Froes and D. Eylon), 107-120; 1984, Warrendale, PA, Metallurgical Society of AIME. powder metallurgy plasma rotating electrode process powder under vacuum (pulverisation sous vide) rotating electrode atomisation rotating electrode process rapid omnidirectional compaction rapid solidification strain energising process thermochemical processing thermochemical treated thermally induced porosity vacuum hot pressing. REA REP ROC RS SEP TCP TCT TIP VHP PM PREP PSV References blended elemental broken up structure computer aided design computer aided inspection computer aided manufacture Crucible ceramic mould cold isostatic press commercially pure continuous shot casting Colt-Crucible hydrogenation electron beam rotating disc extra low chlorine fluid die compaction gas atomisation hydride-dehydride hot isostatic pressing ingot metallurgy isothermal ROC non-destructive evaluation near net shapes pre alloyed BE BUS CAD CAl CAM CCM CIP CP CSC C-T EBRD ELCL FDC GA HDH HIP 1M IsoROC NDE NNS PA International Materials Reviews 1990 Vol. 35 NO.3 24. s. ABKOWITZ and D. M. ROWELL: in 'Titanium 1986, Products and applications', Vol. 2, 816; 1987, Dayton, OH, Titanium Development Association. 25. c. A. KELTO, B. A. KOSMAL, D. EYLON, ,and F. H. FROES: J. Met., 1980, 32, (8), 17-25. 26. S. KRISHNAMURTHY, R. G. VOGT, D. EYLON, and F. H. FROES: in 'Progress in powder metallurgy', Vol. 39, (ed. H. S. Nayar), 603-623; 1984, Princeton, NJ,Metal Powder Industries Federation. 27. F. H. FROES and D. EYLON: in 'Titanium', Supplement to Am. Met. Market, 24 June 1983, 10A-11A, 21A. 28. F. H. FROES and I. E. SMUGERESKY (eds.): 'Powder metal- lurgy of ~itanium alloys'; 1980, Warrendale, PA, Metallur- gical Society of AIME. 29. F. H. FROES and D. EYLON (eds.): 'Titanium net shape technologies~; 1984, Warrendale, PA, Metallurgical Society of AIME. 30. F. H. FROES and D. EYLON: in 'ASM metals handbook', 9 edn, Vol. 7, 164-168; 1984, Metals Park, OH, American Society for Metals. 31. F. H. FROES, D. EYLON, and G. FRIEDMAN: in 'ASM metals handbook, 9 edn, Vol. 7,748-755; 1984, Metals Park, OH, American Society for Metals. -' '32. D. EYLON and F. H. FROES: in 'Titanium alloys in surgical implants', STP 796, (ed. H. A. Luckey and F. Kubli), 43-58; 1983, Philadelphia, PA, American Society for Testing and Materials. 33. D. EYLON, R. E. OMLOR, and F. H. FROES: in 'Titanium '80, Science and technology', Vol. 3, (ed. H. Kimura and O. Izumi), 2205~2213; 1980, Warrendale, PA, Metallurgical Society of AIME . 34. F. H. FROES, D. EYLON, and Y. MAHAJAN: in 'Modern developments in powder metallurgy', Vol. 13, (ed. H. H. Hausner et al.), 523-535: 1981, Princeton, NJ, Metal Powder Industries Federation. 35. D. EYLON, F. H. FROES, D. G. HEGGIE, P. A. BLENKINSOP, and R. W. GARDINER: Metall. Trans., 1983, 14A, 2497-2505. 36. s. W. SCHWENKER, D. EYLON, and F. H. FROES: in 'Materials and processes - continuing innovations', Vol. 28, 436-447; 1983, Covina, CA, Society for the Advancement of Material and Process Engineering. 37. D. EYLON, P. R. SMITH, S. W. SCHWENKER, and F. H. FROES: in 'Industrial applications of titanium and zirconium: Third conference', STP 830, (ed. R .. T. Webster and C. S. Young), 48-65; 1984, Philadelphia, PA, American Society for Testing and Materials. 38. D. EYLON and F. H. FROES: J. Met., 1984, 36, (6), 36-41. 39. I.-P. HERTEMAN, D. EYLON, and F. H. FROES: in 'Titanium, Science and technology', Vol. 1, (ed. G. Lutjering et al.), 303-310; 1985, Oberursel, FRG, Deutsche Gesellschaft fur Metallkunde. 40. I.-P. HERTEMAN, D. EYLON, and F. H. FROES: Powder Metall. Int., 1985, 17, (3), 116-118. 41. R. E. OMLOR, G. L. TEETERS, R. I. BACON, D. EYLON, and F. H. FROES: in 'Microstructural science', Vol. 8, (ed. D. W. Stevens et al.), 297-304; 1980, New York, Elsevier North- Holland. 42. R. G. VOGT, D. EYLON, and F. H. FROES: in 'Industrial applications of titanium and zirconium: Fourth volume', STP 917, (ed. C. S. Young and J. C. Durham), 194-208; 1986, Philadelphia, PA, American Society for Testing and Materials. 43. F. H. FROES and R. G. ROWE: in 'Rapidly solidified alloys and their mechanical and magnetic properties', (ed. B. C. G~essen et al.), Mater. Res. Soc. Symp. Proc., 1986",58, 309-334. 44. F. H. FROES and D. EYLON (eds.): 'Titanium, Rapid solidification technology'; 1986, Warrendale, PA, Metallur- gical Society of AIME. 45. R. G. ROWE and F. H. FROES: Unpublished work. 46. J. C. WILLIAMS, M. F. X. GIGLIOTII, and F. H. FROES: Un- published work. 47. s. H. WHANG: J. Mater. Sci., 1986, 21, 2224. 48. F. D. FROES and s. I. SAVAGE (eds.): 'Processing of structural metals by rapid solidification'; 1987, Metals Park, OH, ASM International. ' Froes and Eylon Powder metallurgy of Ti alloys 179 49. C. SURYANARAYANA, F. H. FROES, and R. G. ROWE: Unpub- lished work. ' 50. F. H. FROES and R. G. ROWE: in 'Proc. 6th world titanium conf.', (ed. P. Lacombe et al.), 801-810; 1989, Paris, Societe Fran 180 Froes and Eylon Powder metallurgy of Ti alloys 76. G. WELSCH,Y.-T. LEE, P. c. ELOFF, D. EYLON,and F. H. FROES: Metall. Trans., 1983, 14A, (4), 761-769. 77. J. T. CAMMETI,D. EYLON,P. R. SMITH, C. M. COOKE,and F. H. FROES: in 'Proc. int. conf. on shot-peening (ICSP-2)', (ed. H. O. Fuchs), 181-193; 1984,. Chicago, IL, American Shot Peening Society. 78. D. EYLON,C. M. COOKE, and F. H. FROES: in 'Microstructural science', Vol. 12, (ed. D. O. Northwood et al.), 445-458; 1985, Metals Park, OH, American Society for Metals. 79. D. EYLON, F. H. FROES, and R. W. GARDINER:J. Met., 1983, 35, (2), 35-47. 80. D. EYLON, F. H. FROES, and R. W. GARDINER: in 'Titanium technology: Present status and future trends', (ed. F. H. Froes et al.), 35-48; 1985, Dayton, OH, Titanium Devel- opment Association. 81. D. EYLON AND F. H. FROES: in 'Titanium net shape technologies', (ed. F. H. Froes and D. Eylon), 155-177; 1984, Warrendale, PA, Metallurgical Society of AIME. 82. J. E. SMUGERESKYand N. R. MOODY:in 'Titanium net shape technologies', (ed. F. H. Froes and D. Eylon), 131-143; 1984, Warrendale, PA, Metallurgical Society of AIME. 83. 1. 1. BONINI and K. K. SANKARAN: in 'Progress in powder metallurgy', Vol. 38, (ed. J. G. Bewley and S. W. McGee), 555; 1983, Princeton, NJ, Metal Powder Industries Federation. 84. R. R. BOYER, D. EYLON, C. F. YOLTON, and F. H. FROES: in 'Titanium net shape technologies', (ed. F. H. Froes and D. Eylon), 63-78; 1984, Warrendale, PA, Metallurgical Society of AIME. 85. R. R. BOYER, D. EYLON, and F. H. FROES: in 'Titanium, Science and technology', Vol. 2, (ed. G. Lutjering et al.), 1307-1313; 1985, Oberursel, FRG, Deutsche Gesellschaft fur Metallkunde. 86. R. R. BOYER,D. EYLON,and F..H. FROES:Powder Metall. Int., 1985, 17, (5),239-240. 87. M. HAGIWARA,Y. KAIEDA,arid Y. KAWABE:in 'Titanium 1986, Products and applications', Vol. 2, 850; 1987, Dayton, OH, Titanium Development Association. 88. v. S. MOXSONand G. I. FRIEDMAN: in 'Progress in powder metallurgy', Vol. 42, (ed. E. A. Carlson and G. Gaines), 489; 1986, Princeton, NJ, Metal Powder Industries Federation. 89. v. S. MOXSONand G. I. FRIEDMAN:Met. Powder Rep., 1988, 43, (2), 88. 90. B. FERGUSON, A. KUHN, O. D. SMITH, and F. HOFSTATIER: Int. Powder Metall., 1984, 20, (2), 131. 91. s. FUJISHIRO:Personal communication, July 1985. 92. 1. V. WOOD:Mater. Des., 1983, 4, 673. 93. D. H. RO and M. w. TOAZ: in 'Progress in powder metallurgy', Vol. 38, (ed. J. G. Bewley and S. W. McGee), 311; 1983, Princeton, NJ, Metal Powder Industries Federation. 94. A. R. E. SINGER:Met. Techno!., 1984, 11, 99-104. 95. A. R. E. SINGER:Met. Mater., 1970, 4, 246-250, 257. 96. A. R. E. SINGER:J. Inst. Met., 1972, 100, 185-190. 97. 1. 1. LUCASand P. P. KONIECZNY:Metall. Trans., 1971,2,911. 98. D. EYLON and c. M. PIERCE: Metal!. Trans., 1976, 7A, 111-121. 99. M. PETERS, A. GYSLER, and G. LUTJERING: in 'Titanium '80, Science and technology', Vol. 3, (ed. H. Kimura and O. Izumi), 1977; 1980, Warrendale, PA, Metallurgical Society of AIME. 100. A. GYSLER, 1. LINDIGKEIT, and G. LUTJERING: in 'Proc. 5th ICSMA Conf.', Aachen, FRG, 1979, Vol. 2,1113. 101. G. R. YODER,L. A. COOLEY,and T. W. CROOKER:in 'Proc. 23rd structures and materials conf.', 132; 1982,· New York, American Institute of Aeronautics and Astronautics. 102. H. I. AARONSON,D. EYLON, C. M. COOKE, and F. H. FIioES: Unpublished work, 1984-89. 103. F. H. FROES, C. M. COOKE,D. EYLON, and K. C. RUSSELL: in 'Proc. 6th world titanium conf.', (ed. P. Lacombe et al.), 1161-1166; 1989, Paris, Societe Fran Manufacturing Technology Division, Wright-Patterson Air Force Base, OH, June 1986. 144. D. B. DAWSON and M. G. ULITCHNY: in 'Powder metallurgy of titanium alloys', (ed. F. H. Froes and J. E. Smugeresky), 115-126; 1980, Warrendale, PA, Metallurgical Society of AIME. 145. J. E. SMUGERESKY and D. B. DAWSON: in 'Powder metallurgy of titanium alloys', (ed. F. H. Froes and J. E. Smuger- esky), 127-138; 1980, Warrendale, PA, Metallurgical Society of AIME. 146. R. H. WITT and w. T. HIGHBERGER: in 'Powder metallurgy of titanium alloys,' (ed. F. H. Froes and J. E., Smugeresky), 255-265; 1980, Warrendale, PA, Metallurgical Society of AIME. 147. R. H. WITT: in 'Materials and processes - Continuing innovations', Vol. 28, 422; 1983, Covina, CA, Society for the Advancement of Material and Process Engineering. 148. c. C. CHEN: in 'Powder metallurgy of titanium alloys', (ed. F. H. Froes and J. E. Smugeresky), 139-149; 1980, Warrendale,PA, Metallurgical Society of AIME. 149. I. WEISS, F. H. FROES, D. EYLON, and c. C. CHEN: in 'Titanium net shape technologies', (ed. F. H. Froes and D. Eylon), 79-94; 1984, Warrendale, PA, Metallurgical Society of AIME. 150. D. W. BECKER, W. A. BAESLACK, III, and F. H. FROES: in 'Powder metallurgy of titanium alloys', (ed. F. H. Froes and J. E. Smugeresky), 217-228; 1980, Warrendale, PA, Metallurgical Society of AIME. 151. w. A. BAESLACK, III, D. W. BECKER, and F. H. FROES: J. Met., 1984, 36, (5), 46-58. 152. F. H. FROES, and w. T. HIGHBERGER: in 'Titanium technology: Present status and future trends', (ed. F. H. Froes et al.), 95-102; 1985, Dayton, OH, Titanium Development Association. 153. N. C. BIRLA, V. DePIERRE, and A. M. ADAIR: Powder Metall., 1978,21, (3), 163-167. 154. R. H. WITT and A. L. FERRERI: in 'Overcoming material boundaries', Vol. 17, 296-305; 1985, Covina, CA, Society for the Advancement of Material and Process Engineering. 155. R. G. VOGT, D. EYLON, R. R. BOYER, and F. H. FROES: in 'Titanium, Rapid solidification technology', (ed. F. H. Froes and D. Eylon), 195-199; 1986, Warrendale, PA, Metallurgical Society of AIME. 156. F. H. FROES, D. EYLON, R. G. ROWE, and c. F. YOLTON: Bull. Mater. Sci., 1989, 12, (3-4), 293-311. 157. B. BORCHERT, D. EYLON, G. WIRTH, and F. H. FROES: Unpublished work, 1987. 158. B. BORCHERT, H. SCHMID, and 1. WORTMANN: in 'Overcoming material boundaries', Vol. 17,27-33; 1985, Covina, CA, Society for the Advancement of Material and Process Engineering. 159. B. BORCHERT, H. SCHMID, and 1. WORTMANN: in 'Titanium, Science and technology', Vol. 1 (ed. G. Lutjering et al.), 295; 1985, Oberursel, FRG, Deutsche Gesellschaft fur Metallkunde. 160. s. FUJISHIRO, F. H. FROES, T. MATSUMOTO, and D. EYLON: in 'Titanium, Science and technology', Vol. 1, (ed. G. Lutjering et al.), 593-600; 1985, Oberursel, FRG, Deutsche Gesellschaft fur Metallkunde. 161. c. F. YOLTON, T. LIZZI, V. K. CHANDHOK, and 1. H. MOLL: in 'Progress in powder metallurgy', Vol. 42, (compiled by E. A. Carlson and G. Gaines), 479-488; 1986, Princeton, NJ, Metal Powder Industries Federation. 162. c. F. YOLTON, T. LIZZI, V. K. CHANDHOK, and 1. H. MOLL: in 'Titanium, Rapid solidification technology', (ed. F. H. froes and D. Eylon), 263-271; 1986, Warrendale, PA, Metallurgical Society of AIME. 163. B. LONDON, S. M. L. SASTRY, C. G. LEVI, and 1. VALENCIA: Presented at 2nd Int. ConL on Rapidly Solidified Materials, San Diego, CA, 1988. 164. K. R. TEAL, A. G. JACKSON, D. EYLON, and F. H. FRC)ES: in 'Titanium, Rapid solidification technology', (ed. F. H. Froes and D. Eylon), 231-238; 1986, Warrendale, PA, Metallurgical Society of AIME. . 165. R. G. ROWE, 1. A., SUTLIFF, and E. F. KOCH: in 'Titanium, Rapid solidification technology', (ed. F. H. Froes and D. Eylon), 239-248; 1986, Warrendale, PA, Metallurgical Society of AIME. 166. s. A. COURT, S. A. SKEWES, M. H. LORETTO, and H. L. FRASER: Froes and Eylon Powder metallurgy of Ti alloys 181 in 'Titanium, Rapid solidification technology', (ed. F. H. Froes and D. Eylon), 249-253; 1986, Warrendale, PA, Metallurgical Society of AIME. 167. c. H. WARD, T. F. BRODERICK, A. G. JACKSON, R. G. ROWE, and F. H. FROES: in 'Processing of structural metals by rapid solidification', (ed. F. H. Froes and S. J. Savage), 243-246; 1987, Metals Park, OH, ASM International. 168. G. VENKATARAMAN, A. G. JACKSON, and F. H. FROES: in 'Rapidly solidified materials: properties and processing', (ed. P. W. Lee and J.B. Moll), 91-97; 1988, Metals Park, OH, ASM International. 169. c. F. YOLTON and 1. H. MOLL: in 'Progress in powder metallurgy', Vol. 43, (compiled by C. L. Freeby and H. Hjort), 49-63; 1987, Princeton, NJ, Metal Powder Industries Federation. 170. R. G. VOGT, D .. EYLON, and F. H. FROES: in 'Titanium, Rapid solidification technology', (ed. F. H. Froes and D. Eylon), 177-194; 1986, Warrendale, PA, Metallurgical Society of AIME. 171. H. D. HANES, D. A. SEIFERT, and c. R. WATTS: in 'Hot isostatic processing'; 1979, Columbus, OH, Battelle Press. 172. w. H. KAO and L. M. ORSBORN: in 'Powder metallurgy of titanium alloys', (ed. F. H. Froes and J. E. Smugeresky), 163-174; 1980, Warrendale, PA, Metallurgical Society of AIME. 173. I. A. MARTORELL, Y. R. MAHAJAN, D. EYLON, and F. H. FROES: Unpublished work, 1985. 174. D. EYLON and F. H. FROES: Met. Powder Rep., 1986,41, (4), 287-293. 175. D. EYLON and F. H. FROES: in 'Titanium, Rapid solidification technology', (ed. F. H. Froes and D. Eylon), 273-289; 1986, Warrendale, PA, Metallurgical Society of AIME. 176. D. EYLON and F. H. FROES: US Patent 4 714 587, Dec. 1987. 177. D. EYLON and F. H. FROES: in 'Proc. 3rd int. conf. on isostatic pressing (IS03)', Vol. 2, 15-1 to 15-26; 1986, London, MPR Publishing Services. 178. c. A. KELTO: in 'ASM metals handbook', 9 edn, Vol. 7, 542; 1984, Metals Park, OH, American Society for Metals. 179. Y. R. MAHAJAN~ D. EYLON, C. A.KELTO, T. EGERER, and F. H. FROES: in 'Titanium, Science and technology', Vol. 1, (ed. G. Lutjering et al.), 339-346; 1985, Oberursel, FRG, Deutsche Gesellschaft fur Metallkunde. 180. Y. R. MAHAJAN, D. EYLON, C. A. KELTO, T. EGERER, and F. H. FROES: Powder Metall. Int., 1985, 17, (2),75-78. 181. Y. R. MAHAJAN, D. EYLON, C. A. KELTO, and F. H. FROES: in 'Progress in powder metallurgy', Vol. 41, (ed. H. I. Sanderow et al.), 163-171; 1986, Princeton, NJ, Metal Powder Industries Federation. - 182. D. EYLON, C. A. KELTO, A. F. HAYES, and F. H. FROES: in 'Progress in powder metallurgy', Vol. 43, (compiled by C. L. Freeby and H. Hjort), 33-47; 1987, Princeton, NJ, Metal Powder Industries Federation. 183. w. KEINATH, R. MOHS, and w. BUNK: in 'Titanium '80, Science and technology', Vol. 3, (ed. H. Kimura and O. Izumi), 2215-2221; 1980, Warrendale, PA, Metallurgical Society of AIME. 184. E. 1. DULlS, V. K. CHANDHOK, F. H. FROES, and L. P. CLARK: in 'Materials synergisms', Vol. 10, 316-329; 1978, Azusa, CA, Society for the Advancement of Material and Process Engineering. 185. R~ F. GEISENDORFER: in 'Powder metallurgy of titanium alloys', (ed. F. H. Froes and J. E. Smugeresky), 151-162; 1980, Warrendale, PA, Metallurgical Society of AIME. 186. c. C. CHEN: J. Met., 1982, 34, (11), 30. 187. H. MARGOLIN and P. COHEN: in 'Titanium '80, Science and technology', Vol. 2, (ed. H. Kimura and O. Izumi), 1555; 1980, Warrendale, PA, Metallurgical Society of AIME. 188. I. WEISS, G. WELSCH, F. H. FROES, and D. EYLON: in 'Titanium, Science and technology', Vol. 3,( ed. G. Lutjering et al.), 1503-1510; 1985, Oberursel, FRG, Deutsche Gesellschaft fur Metallkunde. 189. I. WEISS, G. WELSCH, F. H. FROES, and D. EYLON: in 'Strength of metals and alloys', Vol. 2, (ed. H. J. McQueen et al.), 1073-1078; 1985, Oxford, Pergamon Press. 190. I. WEISS, G. WELSCH, F. H. FROES, and D. EYLON: Metall. Trans., 1986, 17A, 1935-1947. 191. A. GOGlA, D. BANERJEE, and D. c. BIRLA: Trans. Indian Inst. Met., 1983, 36, 200. International Materials Reviews 1990 Vol. 35 NO.3 182 Froes and Eylon Powder metallurgy of Ti alloys 192. D. EYLON, T. L. BARTEL, and M. E. ROSENBLUM: Metall. Trans., 1980, l1A, 1361-1367. . 193. D. EYLON, M. E. ROSENBLUM, and s. FUJISHIRO: in 'Titanium '80, Science and technology', Vol. 3, (ed. H. Kimura and O. Izumi), 1845-1854; 1980, Warrendale, Metallurgical Society of AIME. 194. D. EYLON and M. E. ROSENBLUM: Metall. Trans., 1982, 13A, 322-324. 195. F. H. FROES, D. EYLON, G. WIRTH, K.-1. GRUNDHOFF, and W.-SMARSLY: Met. Powder Rep., 1983, 38, (1), 36-41. 196. F. H. FROES, D. EYLON, G. WIRTH, K.-J. GRUNDHOFF, and w. SMARSLY: in 'Progress in powder metallurgy', Vol. 38, (ed. J. G. Bewley and S. W. McGee), 503-519; 1983, Princeton, NJ, Metal Powder Industries Federation. 197. P. R. SMITH, D. EYLON, S. W. SCHWENKER, and F. H. FROES: in 'Advanced processing methods for titanium', (ed. D. F. Hasson and C. H. Hamilton), 61-77; 1982, Warrendale, PA, Metallurgical Society of AIME. 198. F. H. FROES and D. EYLON: in 'Progress in powder metallurgy', Vol. 37, (ed. J. M. Capus and D. L. Dyke), 279-288; 1982, Princeton, NJ, Metal Powder Industries Federation. - 199. c. M. COOKE, D. EYLON, and F. H. FROES: in 'Microstructural Science', Vol. 14, (ed. M. R. Louthan, Jr et at.), 303-314; 1987, Metals Park, OH, American Society for Metals. 200. D. EYLON and F. H. FROES: US Patent4 534 808, Aug. 1985. 201. L. S. STEELE, D. EYLON, K. R. TEAL, I. WEISS, and F. H. FROES: in 'Proc. 6th world titanium conf.', (ed. P. Lacombe et at.), 1009-1014; 1989, Paris, Societe Fran~aise de Metallurgie. 202. R. F. VAUGHAN, P. A. BLENKINSOP, and P. H. MORTON: in AGARD Conf. Proc., No. 200, 11-1, 1978. 203. D. EYLON, S. W. SCHWENKER, and F. H. FROES: Metall. Trans., 1985, 16A, 1526-1531. 204. J. E. SMUGERI;:SKY, 1. C. WILLIANS, and F. H. FROES: in 'Powder metallurgy of titanium alloys', (ed. F. H. Froes and J. E. Smugeresky), 303-308; 1980, Warrendale, PA, Metallurgical Society of AIME. 205. s. W. SCHWENKER, A. W. SOMMER, D. EYLON, and F. H. FROES: Metall. Trans., 1983, 14A, (7), 1524-1528. 206. F. H. FROES, D. EYLON, and s. w. SCHWENKER: in 'Progress in powder metallurgy', Vol. 41, (ed. H. I. Sanderow et at.), 519-530; 1986, Princeton, NJ, Metal Powder Industries Federation. 207. G. WIRTH, K.-1. GRUNDHOFF, F. H. FROES, D. EYLON, and s. w. SCHWENKER: in 'Horizons of powder metallurgy', Part 2, (ed. W. A. Kaysser and W. J. Huppmann), 931-934; 1986, Freiburg, FRG, Verlag Schmid GmbH. 208. G. WIRTH, K.-1. GRUNDHOFF, F. H. FROES, D. EYLON, and s. w. SCHWENKER: in 'Progress in powder metallurgy', Vol. 42, (compiled by E. A. Carlson and G. Gaines), 605-610; 1986, Princeton, NJ, Metal Powder Industries Federation. 209. s. W. SCHWENKER, D. EYLON, and F. H. FROES: Metall. Trans., 1986, 17A, (2), 271-280. 210. D. EYLON, Y. MAHAJAN, N. R. ONTKO, and F. H. FROES: in International Materials Reviews 1990 Vol. 35 NO.3 'Powder metallurgy of titanium alloys', (ed. F. H. Froes and J. E. Smugeresky), 93-102; 1980, Warrendale, PA, Metallurgical Society of AIME. 211. G. P. SINGH, R. C. MANNING, G. R. LEVERANT, F. H. FROES, and D. EYLON: Metall. Powder Rep., 1984, 39, (9), 504-508, 520. 212. H. KWUN, G. P. SINGH, K. FUJISAWA, F. H. FROES, and D. EYLON: NDT Int., 1984, 17 (2), 73-79. 213. w. A. BAESLACK, III, D. W. BECKER, and F. H. FROES: in 'Titanium '80, Science and technology', Vol. 4, (ed. H. Kimura and O. Izumi), 2369-2378; 1980, Warrendale, PA, Metallurgical Society of AIME. 214. L. PARSONS, 1. BRUCE, 1. LANE, and F. H. FROES: Met. Prog., Sept. 1984, 83-94. 215. N. OSBORNE, D. EYLON, and 1. P. CLIFFORD: 'Evaluation of titanium alloy powder compacted to near net shape', Report AFWAL- TR-88-4250, Final Report, Structural Metals Branch, Wright-Patterson Air Force Base, OH, Feb. 1989. 216. 1. F. KAHLES, M. FIELD, D. EYLON, and F. H. FROES: J. Met., 1985,37, (4),27-35. 217. 1. F. KAHLES, M. FIELD, D. EYLON, and F. H. FROES: in 'Titanium technology: Present status and future trends', (ed. F. H. Froes et at.), 147-155; 1985, Dayton, OH, Titanium Development Association. 218. N. G. TUPPER, 1. K. ELBAUM, and H. M. BURTE: J. Met., 1978, 30, (9), 7-13. 219. F. H. FROES: Int. J. Powder Metall., 1987,23, (4),267-269. 220. c. F. YOLTON and J. H. MOLL: in 'Titanium 1986, Products and applications', Vol. 2, 782-800; 1987, Dayton, OH, Titanium Development Association. - 221. R. H. WITT: in 'Titanium 1986, Products and applications', Vol. 2, 801-815; 1987, Dayton, OH, Titanium Devel- .opment Association. 222. F. H. FROES, D. EYLON, and R. G. ROWE: Unpublished work, 1989. 223. J. H. MOLL: Personal communication, Sept. 1986. 224. c. A. KELTO: Personal communication, Sept. 1986. 225. H. GREWE, J. POTSCHKE, and w. WEIGLIN: in 'Horizons of powder metallurgy', Part 1, (ed. W. A. Kaysser and W. J. Huppmann), 75-78; 1986, Freiburg, FRG, Verlag Schmid GmbH. 226. H. GREWE: Personal communication, Aug. 1986. 227. P. WINKLER: Personal communication, Aug. 1986. 228. F. H. FROES, H. B. BOMBERGER, D. EYLON, and R. G. ROWE: in 'Competitive advances in metals and processes', (ed. R. J. Cunningham and M. Schwartz), Vol. 1, 240-254; 1987, Covina, CA, Society for the Advancement of Material and Process Engineering. 229. G. FRIEDMAN: in 'Comparative advances in metals and processes', (ed. R. J. Cunningham and M. Schwartz), Vol. 1, 222-227; 1987, Covina, CA, Society for the Advancement of Material and Process Engineering. 230. M. W. KEARNS, P. A. BLENKINSOP, A. C. -BAKER, and T. W. FARTHING: Int. J. Powder Metall., 1988, 24 (1), 59. INTERNATIONAL MATERIALS REVIEWS Instructions to Authors International Materials Reviews exists for the publication of concise, critical reviews by internationally accepted experts, of published work relevant to metals, structural ceramics, composite materials, and electronic materials, conforming to the following requirements: - 1. NATURE OF REVIEWS Reviews must be authoritative and convey accurately the current position to those who are not specialists in the par- ticular field, but who are likely to be interested in it. The review has to be critical and interpret the literature covered. Annotated literature surveys, however comprehensive, will not be accepted, and the review author should not as a general rule include previously unpublished work more suit- able for separate publication. The author should try to write the review so that it can be read with minimum difficulty by the non-specialist. 2. SCOPE OF THE REVIEW In the first review of a subject it may be necessary to include material over 10 years old, and, if so, this must be on a very selective basis, the criterion being that exclusion wo~ldseri- ously affect comprehension by the non-specialist reader. Where an adequate and readily available critical, inter- pretative review exists, the author should concentrate on developments subsequent to that review. All the most important items of literature should be sur- veyed and the accompanying bibliography must be compre- hensive, but selective. Where publication of a specific piece of work is duplicated in more than one journal only one reference should be made. From each section of the review one or two key references may be marked with an asterisk in the bibliography as suitable for further reading. It is expected that the author will draw upon personal experience to comment critically throughout the review. 3. LENGTH OF REVIEWS Reviews should always be concise and would normally fall within the range of 3000 to 12000 words. 4. SYNOPSIS A prospective author should submit a synopsis of the proposed review to the Editor of International Materials Reviews giving a brief outline of the contents, together with an indication of the length (in number of words) of each section. The synopsis will then be examined by the Editorial -Panel, and approved, returned for modification, or declined as appropriate. When a synopsis has been approved the author will be asked to submit· the review by an agreed date which will normally be within 12 months. Should the manuscript be delayed for considerably longer than this period the Editorial Panel may wish to commission an alternative author in the interests of the subscribers. This course of action would only be taken after due consultation with the initial author. 5. REFEREES To maintain the high standard of the reviews, all con- tributions - whether invited or submitted on the author's initiative - will be refereed by two experts in the field of the review. 6. PAYMENT OF AUTHORS As a token of appreciation, an honorarium is offered to authors of reviews published. 7. PREPARATION OF REVIEWS Manuscripts ' Manuscripts (two copies are required) should be typed, using double spacing, on one side only of A4 paper. Ample margins should be provided, not less than 40 mm on the left, and other margins not less than 25 mm. Duplicated reports, unless they fulfil these conditions and are printed on non- absorbent paper, are not acceptable. The, typescript should be presented in the order: (i) title page, (ii) synopsis, (iii) text, (iv) references, (v) appendix/ices, (vi) tables, (vii) figure captions, each of which should start on a new page. The pages should be numbered consecutively, with the title page as page 1. Pages added as the result of revision should be given the number of the previous page followed by a, b,etc. (a) Title page The title page should contain the title of the review, the name(s) of the author(s), and the establishmentJs) at which they work. If at the time the review was prepared any of the authors had a different permanent address, then it should be given, as should the present address of any of the authors who have since moved elsewhere. (b) Abstract Each review should be accompanied by a short abstract, of not more than 150 words. This should consist of a concise summary of the content of the review, drawing attention to fundamental principles, recent advances in knowledge or development, and to the present and potential applications to materials knowledge or practice. It should be intelligible in itself without reference to the review: thus it should not refer to illustrations or tables in the review by their numbers, or include bibliographical references. (c) Text Headings Clear differentiation should be made between the headings of sections, subsections, etc. Although the deci- mal system of numbering (§2, §2.1, §2.1.1, etc.) is not used in the journal, its use in typescripts is a valuable guide to the editorial staff. Mathematics and symbols Any mathematical expression or symbol that cannot be typed should be handwritten in black ink as clearly as possible. Any special characters (e.g. script letters, less common symbols) should be identified in the margin at their first appearance. Great care should be taken to distinguish between characters that are easily confused with each other, e.g. 'one' and 'ell' (1, 1), 'zero' and 'oh' (0,0), and capital and lower-case 'cees' and 'esses' (C, c; S, s). If a large number of symbols are used, a 'List of symbols' should be included before the Introduction. Units Authors are explicitly requested to submit' their reviews using SI units. Reviews using other units will be considered by the referees but if accepted for publication the authors will be required to make the conversions to SI. Numbering of equations, tables, figures, and references Equations, tables, and figures should all be numbered serially throughout the text, and referred to in the text, thus: 'equations (1), (2), (3), ... 'Tables 1, 2, 3, ... ', 'Figs. 1, 2, 3, ... '. References should also be numbered serially; the numbers should be typed as superior characters, thus, 1,2 outside any punctuation marks. References cited for the first International Materials Reviews 1990 Vol. 35 NO.3 183 184 Instructions to Authors time in a table or figure caption should be numbered as if they appeared in the text where the table or figure is first mentioned. Footnotes These should be used sparingly. Often footnotes can be incorporatea in the text, within parentheses. ~eally necessary footnotes are best indicated in the manuscnpt by placing them between two horizontal rules immediately following the line to which they refer, and not at the foot of the page. Acknowledgments These should all appear togetlier in a section immediately preceding the references. (d) References The references should be set out in a list, numbered accord- ing to their appearance in the text. All references given should be complete (examples are given below), and should be verified at source by the author(s). The inclusion of confidential, restricted, or internal reports not readily acces- sible to readers should be avoided wherever possible, as should personal communications. Examples of references Journals 1. v. G. RIVLIN: Int. Me tall. Rev., 1984,29,299-327. 2. D. J. WRIGHT: Energy Policy, Dec. 1974,2,307-315. The journal abbreviations closely follow those use~ in Metals Abstracts. If the abbreviation is not known, the Jour- , nal title should be given in full. For journals in which the pagination is not consecutive throughout the volume, i.e. journals for which each issue is numbered afresh from page 1, it is essential to give the month or part number. Books 3. E. C. ROLLASON: 'Metallurgy for engineers', 2 edn, 62-82; 1949, London, Edward Arnold. 4. T. H. C. CHILDS and A. B. SMITH: in 'Towards improved performance of tool materials', (ed. R. S. Irani et al.), 235-239; 1981, London, The Metals Society. Conference proceedings 5. A. IKEDA, T. KANEKO, and F. TERASAKI: in Proc. Conf. 'Corrosion 80', Chicago, Ill., March 1980, National Association of Corrosion Engineers, Paper 8. Report 6. R. D. NICHOLSON: 'Interfacial structures in nickel-based transition joints after long term service', ~eport RDfMfNI131, Central Electricity Generating Board, Marchwood, 1980. (e) Tables Tabular matter should be kept as simple as possible. Each table should have a title, Column headings should be as brief as possible, and the number of columns should be kept to a minimum. International Materials Reviews 1990 Vol. 35 NO.3 (f) Figure captions Each figure should have a caption. For figures consisting of two or more parts, subcaptions should be given separately above the main caption. A caption to a mi.crograph which does not have a scale bar should give the magnification in the form' x 150'. IllUSTRATIONS All illustrations must be clearly numbered - on diagrams in pencil, well away from the main area of the figure, and on photographs lightly on the back (a heavy impression may well be reproduced in printing). (a) Diagrams When a review is submitted, two sets of diagrams should be supplied for refereeing purposes; these diagrams need not be of a quality suitable for reproduction. In general, the Insti- tute cannot undertake to trace or redraw line figures in order to achieve uniformity of style and lettering. Instead, when papers are returned after refereeing, authors are asked to prepare diagrams for reproduction purposes according to instructions marked on one of the sets of figures originally submitted. In preparing diagrams for reproduction authors are asked to use black Indian ink on tracing cloth, Bristol board, or stout drawing paper. These diagrams should normally be at least twice as large as the propqsed finalsize when printed, which is usually 80mm (single column width) or l66mm (page width). Lettering should be of a size that reduces to 2 mm high in the final version. Graphs Except for units, symbols, etc., which should be as else- where, all lettering should be lower case inside the frame and capitals outside. No background grid lines must be used. (b) Photographs Two sets of glossy prints should be provided. One set should be ,unmounted and unlettered; lettering on the other should conform to the requirements set forth above for line diagrams. (c) Previously publ ished illustrations Where an illustration has previously been published else- where in a book or paper, full details should be supplied. It is the author's responsibility to obtain the consent of the earlier publisher. (d) Return of illustrations Original drawings and photographs accompanying papers are normally retained by the Editorial Department for at least one year before being destroyed. They will be returned on request.