Mechanical Properties of High Temperature Vacuum Brazed HSS on Structural Carbon Steel vvith Simultaneous Heat Treatment Mehanske lastnosti visokotemperaturno vakuumsko spajkanih in istočasno toplotno obdelanih spojev V. Leskovšek1, D. Kmetic, B. Šuštaršič, IMT Ljubljana, Slovenija Prejem rokopisa - received: 1996-10-01; sprejem za objavo - accepted for publication: 1996-11-04 The high temperature vacuum brazing process, at the HSS austenitization temperature makes it possible to carry out simultaneousiy the brazing of HSS on structural carbon steel and heat treatment. The advantages of this process are: increased strength of brazed joints and toughness of the part, optimum hardness and cutting edge strength for a given combination vvorking part/cutting tool. The process is economical when used in modern mass production methods, irrespective of the number of metals to be joined and heat treated. The adaptability makes the process so economical. Key vvords: high temperature vacuum brazing, hardness, microstructure, shear strength, tensile strength, vacuum heat treatment Postopek visoko temperaturnega vakuumskega spajkanja v enokomorni vakuumski peči s homogenim plinskim ohlajanjem pod visokim tlakom vodimo v območju avstenitizacije hitroreznih jekel. Prednost tako izdelanih rezilnih orodij je predvsem v doseganju želene žilavosti nosilnega dela iz konstrukcijskega jekla, v doseganju optimalne trdote rezila izdelanega iz hitroreznega jekla ter njegove odpornosti proti otopitvi pri dani kombinaciji del/orodje. Trdnostne lastnosti vezne plasti so odvisne od dodajnega materiala, tehnologije izdelave in pogojev vakuumske toplotne obdelave. Uporaba tega postopka je ekonomična, če moramo spojiti in vakuumsko toplotno obdelati le nekaj ali pa večje število orodij. Ključne besede: visoko temperaturno vakuumsko spajkanje, trdota, mikrostukture, strižna trdnost, natezna trdnost, vakuumska toplotna obdelava 1 Introduction High temperature vacuum brazing is a method of joining of metals by means of heat and filler metal in vacuum at temperatures above 900°C, yet below the melting point of the joined metals, and vvith no use of fluxes. The products are defect-free joints vvith very high bonding strength that can even reach the strength of the joined metal in many cases (e.g. steel, nickel or cobalt alloys). The high temperature vacuum brazing of HSS on structural carbon steel vvith simultaneous heat treatment is performed in single chamber vacuum furnaces, vvith uniform high-pressure gas quenching at the austenitization temperature of HSS. In this vvork high temperature brazed joints of HSS and structural carbon steel vvith simultaneous heat treatment vvere investigated. Tvvo brazing alloys based on Ni-Cr-Si and copper vvere applied as filler metals. The shear strength of an overlap joint and the tensile strength of a but joint as vvell as, the microstructure and fracture surface vvere investigated. The advantages of the process are, the requested toughness of the carrying part from structural carbon steel and the optimum hardness and cutting edge strength of HSS for the given combination of vvorking part/cutting tool. Such mechanical properties of cutting tools 1 Vojteh LESKOVŠEK. dipl.inž. Inštitut za kovinske materiale in tehnologije 1000 Ljubljana. Lepi pot 11. Slovenija manufactured in the conventional way from HSS can on!y be obtained by an additional tempering operation. Other advantages of the high temperature vacuum process are energy savings, the omittance of expensive tool steels and their cleaning, as vvell as, fevv parts are to be joined or hundreds of thousands vvhen it is economical to use vacuum brazing vvith modern mass production methods. The adaptability makes vacuum brazing of in-creasing use in the metal-joining processes. 2 Basic factors affecting the mechanical properties of the brazed joint The strength of the filler metal is one of the main factors influencing the strength properties of the brazing joint, since it is a direct measure for the strength properties of the joints. Therefore, joints brazed vvith nickel-base filler metal are stronger than those brazed vvith cop-per-base filler metal. The narrovv joint clearance causes a high capillary filling pressure; therefore, the gap should be parallel over the vvhole length of the joint. Only in this way by increased capillary filling pressure the filler metal can be aspired into the gap. The most favourable joint clearance for high vacuum temperature brazing is approximately 0 - 100 pm, vvhen measured at the brazing temperature. Figure 1 shovvs schematically the relation betvveen joint clearance and the tensile strength of the joint for flux brazing and high temperature brazing1. O) C 153 (152.87) >40 jL 60 >40 > C; bal. Ni Figure 6: Tensile test specimen vvith but joint Slika 6: Natezni preizkušanec s čelnim spojem To get a higher strength of the joint or to make the fixturing of parts to be brazed easier, a lap joint should be selected. This joint should be designed to obtain the same stability under load of the joint and of the base metal. The lap length is then function of the tensile strength of the base metal and the shearing strength of the joint: U = ^ (1) T vvhere U = length of the lap in mm, Rm = tensile strength of base metal in Nmm"2, t = shearing strength of the joint in Nmm"2, t = thickness of base metal in mm. If, in addition, a safety factor and an impairment of the joint caused by small brazing errors is taken into ac-count, then the length of the lap should be 3 to 6 times the thickness of the base metal. Generally, three times the base metal is sufficient for metals of low tensile strength; six times should be used for metals of high tensile strength1. The but joint is used for thicker parts (t > 2 mm) if a lap joint is not possible1. In contrast to soldering, the sta-bility under load of this type of joint is often sufficient for practical use if the parts are brazed. Experiments3 vvere performed on shear specimens vvith single and fourfold overlap, (Figure 5). The lamel-lae from HSS and structural carbon steels vvere, finely ground after rough machining. Measurements shovved that the surface roughness Ra = 0.44 ^im in the longitudi-nal direetion vvas equal for both surfaces. The test specimen vvith but joint shovvn in Figure 6 vvas used for the tensile test. For the brazing of the shear and tensile test specimens vvith the but joint the clearance of 80 |4m vvas chosen. The brazing temperature vvas 1120°C for specimens brazed vvith the filler metals LM and Cu, and 1160°C for those brazed vvith the filler metal 30. After diffusion heat treatment, the specimens vvere cooled in nitrogen flovv at the pressure under 5 bar abs, and than double tempered at 550°C, (Figure 7). The brazing vvas performed in a vacuum 5 x 10~2 mbar. Shear and tensile specimens vvith but joints vvere used for metalographical and mechanical research. For the investigation of endurance of brazing joint, tvvo paper knives vvith the dimensions of 425x117x10 mm and one knife vvith the dimension 560x117x10 mm, vvere manufactured from HSS W. No. 1.3343 steel and their bearing parts from the steel W. No. 1.7131 (DIN) steel, (Figure8). The filler metal marked LM vvas used for these knives and considering the knives' shape, a lap joint vvith 80 pm clearance vvas chosen. The brazing temperature vvas 1190°C. After diffusion heat treatment, the knives vvere cooled in a nitrogen flovv at a pressure under 5 bar abs, follovved by double tempering at 540°C, (Figure 7). The brazing vvas performed in a vacuum, 5 x 10"2 mbar. 4 Results and discussion 4.1 Mechanical tests Next to the required properties of structural carbon steel and HSS, the most important property is the bond strength betvveen them. Mechanical tests vvere performed on fourteen shear specimens vvith a single and four-fold overlap and length of the lap of 2 to 6 times the thickness of the base metal and three tensile test specimens brazed vvith LM and Cu filler metal. The joint clearance for high temperature vacuum brazing vvas among 50-70 |im for the specimens brazed vvith fillers LM and 30, and 20-50 um for the specimens Breazing temperature Figure 7: High temperature brazing with simultaneous heat treatment process model Slika 7: Model visoko temperaturnega vakuumskega spajkanja z istočasno toplotno obdelavo Figure 8: Paper knives manufactured by high temperature vacuum brazing vvith simultaneous heat treatment process to achieve a hardness of 64 Hrc Slika 8: Noža za rezanje papirja izdelana po postopku visoko temperaturnega vakuumskega spajkanja in istočasno toplotno obdelana na 64 HRc brazed with the copper filler. Data regarding specimens characteristics and the shear strength obtained by the In-stron tensile testing machine are summarised in table 2. Table 2: Specimens characteristics and the shear strength Tabela 2: Karakteristike preizkušancev in strižne trdnostip rekrovnih spojev Sample Filler metal Overlap Length of the lap Shear strength Nmnr2 A/l LM four-fold 3 x t > 30 A/2* LM four-fold 3 x t 27 A/3 LM four-fold 6 x t > 30 A/4* LM four-fold 6 x t 18 A/5 LM single-fold 3 x t > 71 A/6 LM single-fold 2 x t > 210 B/l 30 four-fold 3 x t > 30 B/2* 30 four-fold 3 x t 27 B/3 30 four-fold 6 x t > 20 B/4 30 single-fold 3 x t > 60 C/l Cu four-fold 3 x t > 32 C/2 Cu four-fold 6 x t > 62 C/3 Cu single-fold 3 x t > 66 C/4 Cu single-fold 2 x t > 205 * Samples fractured in bond layer; C/l- the middle lamellae made from W. No. 1.1141, end lamellae made from W.No. 1.3343; C/2 aH lamellae made from W.No. 1.3343, because of gliding in the chucks, there was no destruction of the sample; C/3- ali lamellae made from W. No. 1.1141. Results in table 2, show that rupture of samples, in general, appeared in the structural carbon steel and not in the bond layer, (Figure 9), since the shear strength of brazed joints was greater than the tensile strength of the structural carbon steel. The sample where the middle lamellae were from the steel W. No. 1.1141, was an excep-tion since the fracture appeared simultaneously on both middle lamellae. The shear strength is dependent upon the overlap shape and the lap length. The maximal shear strength was obtained on samples with a single-fold overlap and with the lap length 2 times the thickness of the base met- M_ i* j^Mf^tfr """ ! jr J < .1 .1 J "4 Figure 9: Shear specimens B/l and C/l vvith a four-fold overlap after the tensile test Slika 9: Strižna preizkušanca B/l in C/l s štirikratnim prekritjem po trgalnem preizkusu al. On samples brazed vvith filler metal LM slightly higher values vvere obtained. After vacuum heat treatment that corresponded to austenitizig and tempering temperatures for HSS M15 (AISI), the strength of the tensile test specimen vvith but joint vvas a little lovver than that for structural carbon steel. The fractures propagated mostly vvithin the bond layer and partly also in structural carbon steel and HSS. By tensile tests, the strength of specimens vvith but joint vvas strongly influenced by defects in the bond layer (sample C/8*). During tensile tests vve did not notice any elongation or reduction of area on the samples. Results of tensile tests are presented in table 3. Table 3: Strength of the tensile test specimen with but joints Tabela 3: Natezne trdnosti čelno spajkanih preizkušancev Sample Filler metal Rc (Nmnr2) Rm (Nmm"2) A/8 LM 330 445 C/7 Cu 340 475 C/8* Cu 325 345 * defects in the bond layer After mechanical tests, a metallographical examina-tion vvas performed. On the single or four-fold overlap Figure 10: Initial microcrack area propagating through the eutectic phase is in the microporous regions, sample A/2 Slika 10: Inicial za nastanek mikrorazpok, ki potekajo po eutektični fazi, so mikroporozna mesta, preizkušanec A/2 after hardening after hardening and tempering W. No. 17131 "(DIN) Figure 11: Microstructure of the bond layer in specimen C/7 Slika 11: Mikrostruktura vezne plasti na preizkušancu C/7 specimens, where fractures appeared in the structural carbon steel, only sporadic microcracks were found in the bond layer. On specimens with fracture in the bond layer, areas with microporosity were noticed, vvithout ex-ception, vvhere microcracks initiated. On specimens brazed vvith the fillers LM and 30, the fracture cracks propagated through the eutectic phase of the bond layer, (Figure 10). As mentioned above, the diffusion of carbon from HSS to structural carbon steel took plače; and conse-quently, the microstructure along the bond layer/struc-tural carbon steel consisted of pearlite and bainite. On specimens brazed vvith copper, cracks appeared at the bond layer/structural carbon steel, respectively, (Figure 11). Tensile test specimens fractured in this region, as vvell. Although carbon is not soluble in copper, the diffusion of carbon from HSS throughout the copper bond layer to structural carbon steel cannot take plače, the mi- - i = M2 (AISI) 2 1.5 1 0.5 0 0 0.5 1 2 Distance in mm Figure 12: Vickers microhardness on transition from the bond layer to HSS and structural carbon steel Slika 12: Potek mikrotrdote HV na prehodu iz vezne plasti v hitrorezno in konstrukcijsko jeklo Figure 13: Fracture through an area of eutectic and austenite phase. sample A/8 Slika 13: Prelom preko eutektika in avstenitne faze, preizkušanec A/8 crostructure along the bond layer/structural carbon steel consisted of ferrite and bainite with traces of pearlite. On the paper knife, the microhardness vvas measured across the bond layer to HSS and the structural carbon steel. The diffusion annealing was carried out with the aim to affect hardness at its transition across the bond layer and Figure 12 shows the microhardness profile ob-tained. It shows that the HSS hardness is decreased, while it is increased in the structural carbon steel. The morphology of fracture surfaces is very hetero-geneous. On the specimens brazed vvith the fillers LM and 30, it vvas possible to identify fracture surfaces that propagated in dendrit's area from those propagated in the eutectic phase and in austenite, (Figs 13 and 14). Ductile fracture on specimens brazed vvith copper propagated mostly vvithin bond layer, (Figure 15). Inclu-sions of copper oxide vvere found in the dimples. 4.2 Microstructural characterisation The used filler metals, structural carbon steel and HSS vvere examined by optical and scanning electron microscopy. The microstructure of the W. No. 1.1141 (DIN) structural carbon steel consisted of ferite-pearlite and bainite vvith a hardness of 145 HV10. The microstructure of the W. No. 1.3343 HSS consisted of a matrix of tempered martensite containing small carbide precipi-tates. The size of austenite grains vvas among 17 and 13 SG depending on the austenitization temperature and the hardness 64 HRc. Figure 16 shovvs the microstructure of the bond layer betvveen the HSS and the structural carbon steel on hard- Figure 14: Fracture surface of the specimen B/2 Slika 14: Prelomna površina preizkušanca B/2 Figure 15: Fracture surface of the specimen C/7 brazed vvith Cu Slika 15: Prelomna površina preizkušanca C/7 spajkanega s Cu - v jamicah so vključki bakrovega oksida ened and tempered specimens A/l and B/2. The specimens vvere brazed vvith the fillers LM and 30. In the bond layer polygonal grains formed because of the diffusion during brazing. The diffusion at the HSS/bond layer border seams to be quicker; therefore, more of this phase is found in the bond layer along the HSS. Along the structural carbon steel/bond layer, the bond layer vvas homogenous. The specimen B/2 vvas ex-amined by SEM, (Figure 17). Figure 16: Microstructure of the high temperature brazed and simultaneously heat treated joints of HSS and structural carbon steel, samples A/l and B/2 Slika 16: Mikrostruktura vezne plasti na preizkušancih A/l in B/2 A detailed investigation in EPMA showed that the larger polygonal grains present along the central line of the bond layer were a phase solidification grains rich in Cr, containing also Ni and Si with traces of W, Mo and V. The smaller grains were carbides, (Figure 18). The in-termetallic phase was hard. The measured microhardness was 500-600 HV. The average matrix microhardness was 195 H V. In the microstructure at the HSS/bond layer border, the effects of the diffusion processes were clearly notice-able. In the thin layer of HSS only carbides particles were noticed, martenzite matrix was transformed be- Figure 17: SEM micrograph of high temperature brazed joint, specimen B/2 Slika 17: Mikrostuktura vezne plasti preizkušanca B/2 posneta s SEM Figure 18: Distribution elements in the bond layer, sample B/2 Slika 18: Porazdelitev elementov v vezni plasti na preizkušancu B/2 cause of diffusion into austenite. This microstructure was very similar to that in the bond layer, (Figure 17). At the bond layer/structural carbon steel border, diffusion of Cr, Ni and Si to structural carbon steel oc- curred. The hardened and tempered samples brazed with filler metal LM and 30 showed along this border a thin layer rich in carbon, (Figure 16) vvith microstructure consisting of a small amount of pearlite and bainite. The diffusion of carbon vvas more rapid on the samples brazed vvith the filler LM. The microstructure of W. No. 1.7131 (DIN) struetural carbon steel used for bearing part of paper knives con-sisted of tempered martensite and bainite. Austenite grains vvere coarse, due to the high austenitization temperature. The microstructure of the W. No. 1.3343 HSS consisted of a matrix of tempered martensite containing small carbide precipitates and the size of austenite grains of 14 SG. The microstructure of the bond layer vvas iden-tical as in hardened and tempered samples brazed vvith filler metal LM. 5 Conclusion Mechanical tests and metallographic observations vvere carried on high temperature vacuum brazed and si-multaneously heat treated shear specimens vvith single and four-fold overlap and tensile test specimens vvith but joint. Tvvo Ni-Cr-Si brazed metals as vvell as copper served as filler metal. During the heat treatment, rapid diffusion processes occurred betvveen the liquid and the hard phase, especially along the HSS border. By use of Ni-Cr-Si based filler metal the formation of intermetallic phases, eutectic phases and carbides in the bond layer, and a net of eutectic carbides and voids on the austenite net along the bond layer/HSS border, vvere observed. The mechanical properties of the bond layer depend on specimen design, manufacture and heat treatment conditions. The bond layer must be as thin and as ho-mogenous as possible and must shovvs no porosity or mi-crocracks. Intermetallic phases and carbides cannot be eliminated, due to the speed of the diffusion processes, which are very high on the HSS heat treatment temperature. 6 References 1 J. VV. Bouvvman, High Temperature Vacuum Brazing, lpsen Instruc-tion Manuals 2 V. Leskovšek, D. Kmetič, J. Gnamuš and G. Rihar: High temperature vacuum brazing of HSS on construction steel vvith simultaneous heat treatment, Vuolo, 20, 1990, 2, 512- 515 3 D. Kmetič, V. Leskovšek, J, Žvokelj and J. Gnamuš: Trdnostne lastnosti visokotemperaturno spajkanih spojev v vakuumu, KZT, 27, 1993, 1-2, 69-73