scieee Open visual document viewer

Development of electric resistance sintering process for the fabrication of hard metals: Processing, microstructure and mechanical properties

Lagos, Miguel Ángel; Agote, Íñigo; Schubert, Thomas; Weissgaerber, Thomas; Gallardo Fuentes, José María; Montes Martos, Juan Manuel; Prakash, Leo; Andreouli, D.; Oikonomou, Vivi; López, D.; Calero, J. A.

Abstract

This work presents the development of the Electrical Resistance Sintering (ERS) process for the fabrication of hard metals. The compositions of the materials produced were WC with 6 and 10 wt% of Co. In addition to the specific characteristics of the technology, the characterization of the produced parts is presented and compared to materials obtained by conventional processes. The parts produced by ERS present densities comparable to the ones obtained by conventional methods. The microstructural comparison shows a considerable grain size reduction in the ERS materials which consequently brings a hardness increase. ERS materials show similar fracture toughness to conventional ones. The very fast sintering allows performing the process without any protective atmosphere, therefore making this process very attractive for the production of materials that need to be sintered under non-oxidising environments. The total duration of the cycle, including heating, holding time and cooling is few seconds. Finally, some considerations about the scale up and possible industrialization of the technology are explained.

Full text

Accep ed Manusc ip De elopmen o elec ic esis ance sin e ing p ocess o he ab ica ion o ha d me als: P ocessing, mic os uc u e and mechanical p ope ies M.A. Lagos, I. Ago e, T. Schube , T. Weissgae be , J.M. Galla do, J.M. Mon es, L. P akash, C. And eouli, V. Oikonomou, D. Lopez, J.A. Cale o PII: S0263-4368(17)30055-0 DOI: doi: 10.1016/j.ij mhm.2017.03.005 Re e ence: RMHM 4426 To appea in: In e na ional Jou nal o Re ac o y Me als and Ha d Ma e ials Recei ed da e: 31 Janua y 2017 Re ised da e: 7 Ma ch 2017 Accep ed da e: 12 Ma ch 2017 Please ci e his a icle as: M.A. Lagos, I. Ago e, T. Schube , T. Weissgae be , J.M. Galla do, J.M. Mon es, L. P akash, C. And eouli, V. Oikonomou, D. Lopez, J.A. Cale o , De elopmen o elec ic esis ance sin e ing p ocess o he ab ica ion o ha d me als: P ocessing, mic os uc u e and mechanical p ope ies. The add ess o he co esponding au ho was cap u ed as a ilia ion o all au ho s. Please check i app op ia e. Rmhm(2017), doi: 10.1016/j.ij mhm.2017.03.005 This is a PDF ile o an unedi ed manusc ip ha has been accep ed o publica ion. As a se ice o ou cus ome s we a e p o iding his ea ly e sion o he manusc ip . The manusc ip will unde go copyedi ing, ypese ing, and e iew o he esul ing p oo be o e i is published in i s inal o m. Please no e ha du ing he p oduc ion p ocess e o s may be disco e ed which could a ec he con en , and all legal disclaime s ha apply o he jou nal pe ain. b ough o you by COREView me ada a, ci a ion and simila pape s a co e.ac.uk p o ided by TECNALIA Publica ions ACCEPTED MANUSCRIPT DEVELOPMENT OF ELECTRIC RESISTANCE SINTERING PROCESS FOR THE FABRICATION OF HARD METALS: PROCESSING, MICROSTRUCTURE AND MECHANICAL PROPERTIES M.A. Lagos1, I. Ago e1, T. Schube 2, T. Weissgae be 2, J.M. Galla do3, J.M. Mon es3, L. P akash4, C. And eouli5, V. Oikonomou5, D. Lopez6, J.A. Cale o6 1 TECNALIA RESEARCH & INNOVATION 2 F aunho e Ins i u e o Manu ac u ing Technology and Ad anced Ma e ials IFAM, B anch Lab D esden 3 Uni e si y o Se ille 4 KYOCERA UNIMERCO TOOLING AS 5 MIRTEC 6 ALEACIONES DE METALES SINTERIZADOS (AMES) ABSTRACT This wo k p esen s he de elopmen o he Elec ical Resis ance Sin e ing (ERS) p ocess o he ab ica ion o Ha d Me als. The composi ions o he ma e ials p oduced we e WC wi h 6 and 10 w % o Co. In addi ion o he speci ic cha ac e is ics o he echnology, he cha ac e iza ion o he p oduced pa s is p esen ed and compa ed o ma e ials ob ained by con en ional p ocesses. The pa s p oduced by ERS p esen densi ies compa able o he ones ob ained by con en ional me hods. The mic os uc u al compa ison shows a conside able g ain size educ ion in he ERS ma e ials which consequen ly b ings a ha dness inc ease. ERS ma e ials show simila ac u e oughness o con en ional ones. The e y as sin e ing allows pe o ming he p ocess wi hou any p o ec i e a mosphe e, he e o e making his p ocess e y a ac i e o he p oduc ion o ma e ials ha need o be sin e ed unde non- oxidising en i onmen s. The o al du a ion o he cycle, including hea ing, holding ime and cooling is ew seconds. Finally, some conside a ions abou he scale up and possible indus ializa ion o he echnology a e explained. Keywo ds: Elec ical sin e ing, ERS, FAST, cu ing ools, ungs en ca bide 1. INTRODUCTION ECAS (elec ic cu en assis ed sin e ing) ga he s a amily o consolida ion me hods in which mechanical p essu e is combined wi h elec ic and he mal ields o enhance pa icle bonding and densi ica ion. The p ima y pu pose o imposed elec ic cu en is o p o ide he equi ed amoun o esis i e hea . The exp ession ield-assis ed sin e ing echniques (FAST) ha e become also popula in li e a u e o denomina e hese echnologies. Indeed, he speed o he p ocesses is he mos ema kable cha ac e is ic and he common ac o among hese echniques. ECAS echniques can be classi ied wi h espec o he p ocessing ime [1]. In con en ional ECAS p ocesses, like SPS (Spa k Plasma Sin e ing) [2], low cu en densi ies and ol age a e applied ( ypically < 10 V and 1 kA/cm2). Wi h hese condi ions, p ocessing ime is in he ange o minu es and o ha eason a con olled a mosphe e is needed. Howe e , in e y as p ocesses like EDS (Elec o Discha ge Sin e ing) [3] and ERS (Elec ic Resis ance Sin e ing) [4-6] highe cu en densi ies a e applied ( ypically > 5 kA/cm2) and he p ocessing ime is ew seconds. The sho du a ion o he cycle pe mi s he p ocessing in ai wi hou any p o ec i e a mosphe e. This is a e y impo an ad an age om he economic poin o iew. In EDS, he elec ic discha ge is p oduced wi h a high- ol age ACCEPTED MANUSCRIPT ACCEPTED MANUSCRIPT capaci o bank, whe eas in ERS he cu en is p oduced by a low- ol age ans o me (a ound 10 V). The e o e, he cu en applied o he sample is easie o con ol in ERS and his imp o es he homogenei y o he p oduc s. Elec ical esis ance sin e ing (ERS), was al eady desc ibed in 1933 by Taylo [7] and la e modi ied by C eme [8] in 1944. Howe e , up o now, he ERS p ocess has been used o he consolida ion o high conduc i e ma e ials (pu e me als) a lab scale, ob aining e y small pieces [4-6]. The de elopmen o medium ol age machines (40 V) opens he possibili y o ob aining la ge pa s and using ma e ials wi h lowe elec ical conduc i i y like composi es and ce me s. WC–Co cemen ed ca bides ha e been widely used o cu ing ools o a ious ma e ials and o he machine pa s which a e equi ed o show high esis ance o ic ional wea . Thei mechanical p ope ies can be modi ied o e a b oad ange by changing he con en o he binding Co phase and he WC g ain size [9-13]. Cemen ed ca bides a e usually p oduced by sin e ing wi h he pa icipa ion o a liquid cobal phase. Howe e , he p esence o his phase du ing he WC–Co sin e ing, s imula es he g ow h o he WC g ains. In gene al, wi h dec easing WC pa icle size p ope ies such as ha dness, wea esis ance, and ans e se up u e s eng h o he composi e a e imp o ed [14-16]. The e ec o he nanos uc u e in he WC-Co ca bides p ope ies has been ex ensi ely s udied [17- 19]. I is known ha ac u e oughness dec eases wi h inc easing ha dness in con en ional composi es, whe eas he inc ease o ha dness in nano-s uc u ed composi es does no u he educe hei bulk ac u e oughness. This implies ha di e en dominan oughening mechanisms exis in he con en ional and nanos uc u ed composi es. Ve y as sin e ing p ocesses o e a unique oppo uni y o a oid he liquid phase sin e ing and hus limi he g ain g ow h. This pape p esen s he de elopmen o he ERS echnology o he p ocessing o ha d me als. The new pilo plan scale equipmen is shown. An elec ic p ess o 15 was combined wi h medium ol age ans o me s in a ully au oma ed ERS equipmen . Mic os uc u al and mechanical cha ac e iza ion o he ma e ials a e p esen ed and compa ed o comme cial sin e -hipped ca bides. In addi ion, some impo an aspec s o he echnology a e desc ibed. 2. MATERIALS AND METHODS WC-Co powde s (submic on size) we e used wi h he ollowing composi ion: 6 w % o Co and 10 w % o Co. G anula ed powde s om a comme cial sou ce we e p ocu ed wi h o ganic wax and a p e- ea men was pe o med in o de o elimina e o ganic componen s. The sphe ical mo phology o he g anules wi h a size be ween 100-200 µm can be obse ed in Figu e 1A. Each g anule is composed o e y ine powde (size less han 1 mic on). The su ace o one g anule is obse ed in Figu e 1 B. Appa en densi y o he g anula ed powde was 3,7 g/cm3. A B Figu e 1. WC-6Co powde , A) SEM image o he g anula ed powde , magni ica ion X200, B) SEM image o he su ace o he g anules, magni ica ion X20000. ACCEPTED MANUSCRIPT ACCEPTED MANUSCRIPT Fo he ERS p ocessing, he powde was illed in an alumina based ce amic die be ween wo coppe elec odes (see he diag am o he ERS machine in Figu e 2A). The dimensions o he p oduced pa s we e 22 mm in diame e wi h a hickness o 10-14 mm. The maximum applied cu en densi y was be ween 4-5 kA/cm2 wi h a holding ime o 500 ms. Maximum load was 100 MPa. An image o he p o o ype machine used in his wo k is p esen ed in Figu e 2B. This ully au oma ed ERS machine was de eloped wi hin he EU unded EFFIPRO p ojec . Maximum ol age o he machine is 40 V wi h a maximum cu en o 35 kA. The p essu e is applied using a 15 ones elec ical p ess. A B Figu e 2. A) Diag am o he echnology, B) P o o ype machine, de eloped wi hin he EFFIPRO p ojec Densi y was measu ed by A chimedes me hod using a Me le AE 240 weigh balance and po osi y was analysed acco ding o he s anda d UNE-EN ISO 4605:1978. Mic os uc u e and semi- quan i a i e chemical composi ion we e analysed by op ical and SEM mic oscopy (Zeiss N ision 40 and Zeiss EVO50). Ha dness Vicke s (HV30) was measu ed acco ding o he s anda d UNE-EN ISO 6507-1:2006. KIC was calcula ed om he leng h o he adial c acks o igina ed in he co ne s o he Vicke s inden a ions acco ding o he o mula p oposed by She y e al. [20]. The analysis o he magne ic coe ci i y HC and sa u a ion magne isa ion MS was pe o med acco ding o DIN ISO 3326 using a Koe zima CS 1.096, om Fö s e GmbH. The W con en o he Co binde was calcula ed om measu emen s o magne ic momen [21]. Measu emen o la ice pa ame e o he Co binde phase was made wi h an X- ay di ac ome e using CuKα adia ion on he su ace o es -pieces which we e elec oly ically e ched o emo e WC. The elec oly e consis ed o 4-mol solu ion o NaOH in eagen g ade. 3. RESULTS AND DISCUSION 3.1. Cha ac e is ics o he ERS p ocessing o he ab ica ion o WC-Co Figu e 3 shows he clea ela ionship be ween he cu en applied and he global densi y ob ained in he ERS samples, measu ed by he A chimedes me hod. Inc easing he densi y o cu en , he global densi y o he samples is inc eased. The ma e ial wi h 10 % o Co p esen ed be e esul s o densi ica ion compa ed o he ma e ial wi h 6 % o Co. ACCEPTED MANUSCRIPT ACCEPTED MANUSCRIPT Figu e 3: Rela ionship be ween he densi y and he elec ic cu en applied (densi y includes he ex e nal po ous laye ; co e o he samples was ully dense) The ERS samples ypically consis o a dense co e su ounded by a po ous su ace laye as i is shown in Figu e 4A. The densi y alues p esen ed in igu e 3 show alues be ween 90 and 95% o he heo e ical densi y due o he ac ha he ou e po ous laye is included. I has been obse ed ha inc easing he cu en densi y, he a ea o he ex e nal po ous laye dec eases. The hickness o he ou e po ous laye in con ac wi h he me allic punches ( op and bo om o he sample) is hinne han ha in con ac wi h he ce amic die walls (see igu e 4B and 4C). A e emo ing his laye , samples ob ained using ERS show e y high densi y, wi hin ypical equi ed alues o ha d me al ools. Po osi y a he co e is e y low (A02 acco ding o he s anda d UNE-EN ISO 4605:1978). A ACCEPTED MANUSCRIPT ACCEPTED MANUSCRIPT B C Figu e 4. C oss sec ion o a pa p oduced by ERS, dis ibu ion o he po osi y: A) Whole sample, B) Ex e nal po osi y, con ac wi h he punches (scale ba 100 mic ons), C) Ex e nal po osi y, con ac wi h he ce amic die (scale ba 100 mic ons) In ERS p ocess, he elec ical cu en passes ac oss he ma e ial and he hea ing is p oduced by joule’s e ec . Tempe a u e in he ma e ial depends on di e en ac o s. One o he mos impo an ones is he g een densi y dis ibu ion o he pa s. I is known ha in a double sided p essing, g een densi y o he pa s is lowe a he co e. Whe e po osi y is highe , elec ical esis ance is also highe . This ac p oduces highe empe a u e a he co e o he ma e ial du ing esis ance hea ing. Ano he impo an ac o is he hea losses p oduced by conduc ion om he ma e ial o he dies and elec odes; his also inc eases he empe a u e g adien om he co e o he pa s o he bo de s. Fo ha eason, ex e nal po osi y is p oduced by he as cooling o he ma e ial in con ac o he coppe punches and ce amic die. Based on p elimina y expe imen al da a, i seems ha i is possible o imp o e his phenomenon educing he ex e nal po osi y by using ce amic dies wi h lowe he mal conduc i i y. 3.2. Mic os uc u al cha ac e iza ion o he p oduc s Mic os uc u es o he ma e ials ob ained by ERS a e shown in Figu e 5 (6 and 10 % Co). In bo h cases mic os uc u e is composed o polygonal WC g ains su ounded by Co me allic ma ix (da ke in he images). Figu e 5. HRSEM images o polished c oss sec ions o ERS/WC6Co (le ) and ERS/WC10Co. In addi ion, Figu e 6 p esen s o compa ison wi h con en ionally sin e ed ma e ial ion-e ched c oss sec ions o ma e ials wi h 6 % o Co ob ained by ERS and Sin e HIP, using in bo h cases he same s a ing powde s. A B Figu e 6. HRSEM images o polished and ion-e ched c oss sec ions o ERS/WC6Co (A) and Sin e HIP ma e ial (B) ACCEPTED MANUSCRIPT ACCEPTED MANUSCRIPT C ys al g ain size was es ima ed by he linea in e cep s me hod (based on DIN EN 623-3). This me hod is based on he numbe o a e age leng hs o in e cep s h ough each c ys al/g ain along a line d awn ac oss he ma e ial su ace. The measu ed leng hs o in e cep s o WC and he Co binde o WC6Co a e gi en in Table 1. The eal phase sizes a e abou 1.4x la ge han he measu ed in e cep s owing o he measu ing p ocedu e. The measu emen s o phase sizes we e done a e an ion e ching o he polished sample o imp o ing he image quali y. The measu ed leng hs o in e cep s o WC gi e a WC-g ain size o abou 274nm o ERS and 326nm o he con en ional p ocess. The eason o his e y low g ain g ow h a e he e y sho p ocessing ime and e y as cooling a e du ing he ERS p ocess. The size o he Co phase is simila in he con en ional and ERS p ocesses. Table 1. Compa ison o phase sizes o di e en ly p ocessed ha dme al WC6Co P ocess Phase Leng h o in e cep s (nm) Numbe o measu ed egions d10 d50 d90 ERS WC 73 196 441 934 Co 28 73 218 310 Sin e -HIP WC 89 233 528 932 Co 28 77 195 417 Compa ing hese esul s wi h p e ious li e a u e, g ain size ob ained in ERS samples is simila o he one epo ed by Spa k Plasma Sin e ing o nanoc ys alline powde s [22]. I is impo an o ake in o accoun ha in he p esen wo k submic on powde s we e used ins ead o nanome ic ones. The e y low g ain g ow h obse ed du ing p ocessing opens he possibili y o nano-c ys alline ha d me als. The con igui y o ERS WC6Co was measu ed wi h abou 67% WC/WC combined wi h 33% WC/Co bounda ies. The co esponding alues o con igui y o he Sin e -HIP samples we e 56% WC/WC and 44% WC/Co. Con igui y dec eases wi h inc easing binde con en and is also dependen on he p ocessing his o y o he ca bide [23-24]. I is di icul o de e mine he e ec o con igui y on he mechanical p ope ies due o he in luence o o he mic os uc u al cha ac e is ics; howe e , i is belie ed ha ha dness inc eases wi h inc easing con igui y [24]. Rega ding he c ys allog aphic phases, Figu e 7 p esen s he XRD pa e ns o samples wi h 10 % o Co. ACCEPTED MANUSCRIPT ACCEPTED MANUSCRIPT Figu e 7. XRD pa e n, samples ob ained by ERS, WC -10Co The e ealed main phases a e WC-hexagonal and Co-cubic; aces o Co-hexagonal we e also de ec ed in bo h samples. Measu emen s o la ice pa ame e o he Co binde phase we e made on he su ace o es -pieces which ha e been elec oly ically e ched o emo e WC. F om he XRD measu emen s la ice pa ame e s o Co we e calcula ed and he concen a ion o W and C in he binde we e es ima ed. The la ice pa ame e o he cc phase can be exp essed as [21]: a = ao+ Q·mw + R·mc whe e ao, Q and R a e cons an s wi h alues o 0.3548 nm, 0.00036 and 0.0012, espec i ely. The a omic pe cen ages o W and C a e gi en by mw and mc espec i ely. Ob ained alues a e p esen ed in Table 2. 3.3. Mechanical p ope ies Ma e ials ob ained by ERS (6 and 10 % Co) p esen ed highe ha dness compa ed o con en ional ma e ials ob ained by Sin e -HIP (see Table 2). A ha dness inc ease o he 5 % was obse ed o he 6 % Co and and inc ease o he 8 % o he 10 % Co. Table 2. Summa y o p ope ies o di e en ly p ocessed ha d me als Compo- si ion P ocess Densi y (g/cm3) Ha dness HV30 F ac u e oughness K1C (MPa√m) HC (Oe) MS (10-7 T*m3*kg-1) Co binde composi ion (w %) W Co binde composi ion (w %) C WC6Co ERS 14.7 1960±15 9.6±0.5 470 101 9.8 0.04 Sin e - HIP 14.8 1860±15 9.5±0.5 385 103 8.8 0.03 WC10Co ERS 14.3 1750±20 10.3±0.5 420 150 9.0 0.14 Sin e - HIP 14.4 1620±15 10.2±0.5 320 154 13.9 ~0 ACCEPTED MANUSCRIPT ACCEPTED MANUSCRIPT In ERS ma e ials, a small a ia ion o he ha dness was obse ed be ween he co e and he su ace o he samples, p esen ing a small inc ease o he ha dness a he su ace. This phenomenon is no comple ely clea a he momen . This can be linked o a sligh ly highe g ain size obse ed a he co e o he samples. Samples a e hea ed by Joule’s e ec du ing ERS and he empe a u e a he co e o he samples is sligh ly highe han a he su ace. This longe ime a high empe a u e can p oduce a e y small inc ease in g ain size. Ano he possible explana ion is he mig a ion o Co om he co e o he samples; his mig a ion can p oduce small mic o-po osi y and he local dec ease o he ha dness. Howe e , no signi ican a ia ions in he Co con en we e obse ed. The applicabili y o he Hall-Pe ch equa ion o ha dness p edic ions: H = E1 + F1·dWC-1/2 has been examined o ma e ials wi h nominally 6 w % Co (app oxima ely 10 ol%). H is he ha dness and dwc he g ain size (WC a i hme ic mean linea in e cep ). E1 and F1 a e cons an alues. Leas squa es eg ession analysis was used o i a linea plo o he da a wi h alues o he cons an s E1 and F1 o 913 and 601, espec i ely. The ha dness measu ed on he ERS samples does no mee he men ioned co ela ion (see Figu e 8); he p edic ion should p oduce a highe ha dness as measu ed based on an in e cep o 200nm. The e o e, i is no clea , i his ela ionship be ween ha dness and WC g ain size can u he be ex ended o he ul a ine g ades. Figu e 8. Rela ionship be ween ha dness and g ains size. ERS ma e ials ma ked as ed ci cles [21] F ac u e oughness (K1c) e eals simila beha iou o he di e en ly p ocessed samples (con en ional and ERS) wi h alues in he ange o 9-10 MPa m0.5 (see able 2). I is known ha ac u e oughness dec eases wi h inc easing ha dness in con en ional composi es, whe eas he inc ease o ha dness in nano-s uc u ed composi es does no u he educe hei bulk ac u e oughness [17-19]. The e y sho p ocessing ime used in ERS (in he o de s o ew seconds) helps con olling he g ain g ow h and hus ob aining ine mic os uc u es. I is belie ed ha his educ ion in g ain size inc eases he ha dness bu main aining he ac u e oughness cons an . The mic os uc u al cha ac e isa ion o he WC–Co ha dme als is o en pe o med using magne ic measu emen s. By measu ing he magne ic sa u a ion MS o he coe ci i y HC, i is possible o ob ain an app oxima ion o he WC g ain size, cobal con en and e en de e mine he p esence o addi ional phases. Gene ally, an inc ease o coe ci i y is obse ed wi h a dec ease in WC g ain size. The compa ison o he measu ed HC alues o ERS wi h Sin e -HIP samples con i ms he educ ion in g ain size obse ed in ERS ma e ials. ACCEPTED MANUSCRIPT