Joining of CrNi steel and AlMg alloy without interlayers/CrNi-terase ja AlMg-sulami vahekihita difusioonkeevitus.

Dahms, Steffen ; Kulu, Priit ; Veinthal, Renno 等

1. INTRODUCTION

The manufacture and application of innovative dissimilar material joints with specific characteristics require continuous development of the joining technology. Presently, however, there is a predominance of similar material joints. Comprehensive investigations on the substance-to-substance joining of dissimilar materials with dissimilar characteristics still have to be conducted.

Diffusion joining in the solid state was used as a joining technology for dissimilar material joints. The joints were created with and without interlayer materials below the melting temperature of the joining partners.

Composite constructions have become a focal point in the optimization of lightweight constructions regarding their stability, due to the fact that in this way the characteristics of dissimilar materials can be optimally exploited. The joining of dissimilar materials, e.g. CrNi steels and AlMg alloys poses a special challenge in the field of joining technology. The following requirements on the joining are to be met:

--manufacture of substance-to-substance joints,

--reduction of and avoiding the creation of intermetallic phases,

--balancing of different melting temperatures,

--reduction and absorption of stress in the joint.

A distinction between similar and dissimilar materials is made in order to characterize joined components. Similar materials are structural steels with different yield stresses. Dissimilar materials are, e.g., steels with light or heavy metals such as the combination of steel and aluminium [1,2].

Diffusion joining is currently a very promising joining procedure to manufacture dissimilar material joints without creating a mutual melt. That is why the creation of intermetallic phases from the melt can be prevented within dissimilar material joints. Such phases influence the strength in the material joint [3j. Besides intermetallic phases, also various material properties have an impact on the joint strength. With dissimilar material joints these are:

--thermal expansion,

--heat conductivity,

--melting temperature,

--porosity and hardness.

One important criterion in the diffusion joining of dissimilar materials is the change of properties of the materials in the joining zone. Gradual changes to the properties occur on boundary surfaces. They strongly depend on the diffusion joining parameters (joining temperature, joining force and joining time). Figure 1a shows the change of properties of dissimilar materials before the diffusion joining process. After the joining process the changes are no longer sudden. Examinations have shown that the properties change gradually, exponentially or parabolically (Fig. 1b). The creation of a diffusion zone leads to a material concentration balancing depending on the joining parameters. This results in a partial balancing of property changes in dissimilar material joints. Stress in the joint is reduced, which leads to a higher joint strength [4,5j.

Dissimilar material joints of ceramics, metals and monocrystals with and without interlayers are studied in [6-8]. Dissimilar material joints of CrNi steel and AlMg alloy are under examination for technical applications in the field of nuclear technology. The joining zone must not be influenced by additional alloy elements in order to guarantee a high safety standard. A substance-to-substance joining without interlayers is necessary in order to prevent material failure in the joining zone. Diffusion welding provides an innovative approach where the joint is created through solid state diffusion as well as dislocations in the crystal structure. Joining in the solid state and optimization of joining parameters enables the achievement of joint strength and ageing stability similar to the base material. This is, however, a precondition in order to meet the high quality requirements for applications in the aerospace industry or for material joints in nuclear technology.

2. EXPERIMENTAL

The material properties are essential for the definition of the process parameters in the diffusion joining in order to manufacture high-quality dissimilar material joints (Table 1).

The lowest melting temperature of the joining component AlMg3 (below 620[degrees]C) defines the joining temperature. Matching face centred cubic (fcc) space lattices are advantageous in order to activate diffusion processes in real crystals along dislocations or grain boundaries. The diffusion is isotropic for joining partners with cubic elementary cells irrespective of their orientation. Additionally, vacancy mechanism and dislocation density are the primary joining mechanisms with metals.

The number of primary slip systems in the face-centred cubic lattice (Fig. 2) is limited to 12 in one elementary cell. Each of the 4 slip planes has 3 slip directions, along which the stress, necessary for a deformation, is the lowest. Slip planes are those with the highest number of atoms. Elastic as well as plastic deformations initially occur along these slip directions. Thus a material, which is under pressure during the diffusion process, is elastically deformed along these directions. However, the deformation ability is limited by the size of the interstitials.

[FIGURE 2 OMITTED]

A thermal process has to take place in order to initiate diffusion processes between the joining partners. The joining force triggers a forced diffusion. The diffusion in metals is a process where atoms are moved due to concentration differences with a tendency to balance them.

The joining experiments were carried out in a high temperature graphite furnace with integrated press (Fig. 3). The dimensions of the test pieces were 20 x 20 x 5 mm, which corresponds to a joining surface of 400 [mm.sup.2]. A joining force of 2000 N equates to a joining pressure of 5 MPa. The heating and cooling rates were 10 K x [min.sup.-1] in a pre-vacuum atmosphere at about [10.sup.-1] mbar. Joining time and temperature were varied (Table 2). Solid dissimilar material joints could be manufactured as a result of the joining experiments. The joint strength and form stability strongly depended on the joining temperature and joining force due to the low hot compressive strength of the AlMg alloy.

[FIGURE 3 OMITTED]

3. RESULTS AND DISCUSSION

Expansion and contraction of the joining parts in a thermal process define the state of stress in the joint. Metals can absorb stress due to their ductile behaviour.

Examination on dissimilar material joints showed that, depending on the material behaviour, different expansions of up to [DELTA][alpha] = 10 x [10.sup.-6] [K.sup.-1] can be compensated. Figure 4 illustrates the expansion progression of the dissimilar material joint of CrNi steel and AlMg alloy from room temperature to joining temperature. Expansion coefficients CTE were measured by means of an Absolut-Dilatometer Linseis L 75/20 under air with a heating and cooling rate of 5 K x [min.sup.-1]. With the above-mentioned material joint the expansion coefficients differ from 6 to 8 x [10.sup.-6] [K.sup.-1] up to the joining temperature range. No hysteresis effects in the expansion behaviour could be observed after the measurements. Depending on the temperature, expansion and contraction are almost constant, which permits to join the surfaces and minimizes the creation of cracks.

[FIGURE 4 OMITTED]

The surfaces of the joining parts or their atoms have to come as close as 10-6 to 10-7 mm in order to achieve atomic interaction. In reality this closeness is impeded by microroughnesses, form imperfections and layers of dirt and/or adsorption layers. All types of form deviations, which restrict the number of possible bonds in the joining zone, are reduced in the activation phase, i.e., physical interactions occur, in order to manufacture a high-quality joint. This can be achieved by surface pre-treatment or cleaning. At the same time the surface is extremely altered depending on the degree of deformation, which also affects pollution layers on the surface as well as layers near the surface. The procedures are very complex and the joining parameters are mutually linked.

The joining surfaces were mechanically ground and polished with a diamond paste in order to meet these requirements. The surface analysis was carried out by means of an autofocus sensor AF16, made by OPM GmbH Karlsruhe, coupled with an optical measuring system and it was used for optical length and profile measurement (topography illustration). Manufacturing quality and final state of the joining surfaces were measured (Table 3).

The mean roughness Ra and the arithmetical mean deviation of the filtered roughness profile over the measurement range were analysed, resulting in [R.sub.a] [less than or equal to] 0.225 gym, which meets the surface requirements for the diffusion joining of metals.

The joining and diffusion zones were characterized by means of SEM examinations. All the tests showed the same characteristic. Closed and flawless diffusion zones were achieved. Figure 5 shows the joint at a temperature of 500[degrees]C and a joining time of 12 h. The material structure of the CrNi steel is compact, the structure of the AlMg alloy, however, is porous. The surfaces were homogeneously joined and a diffusion zone can be seen. The results of the joining tests showed no cracks parallel to the joining zone due to the different expansion behaviour. The ductile behaviour of the joining partners compensated the stress in the joining zone.

EDX-analysis in the diffusion zone or in neighbouring areas enables a qualitative examination of the diffusion behaviour of dissimilar material joints. Figure 6 illustrates the element distribution over the joining zone. The analysis showed that the concentration difference of the alloy elements forms a step in the measurement area. The left side of the diagram shows the CrNi steel with Fe, Cr, Si, Ni and a low Mn proportion. The right side of the diagram shows the AlMg alloy with high Al and low Mg proportion. The concentrations on the left (CrNi steel with alloy elements) are nearly constant. The concentration progressions from maximum to minimum exhibit a diffusion range x of over 100 [micro]m.

Distribution of the joining parameters helps to optimize the joining process and to meet the requirements of the practice. The development of the diffusion areas x of the dissimilar material joints depends on the joining pressure, joining time and joining temperature, as shown in Fig. 7. The diffusion distance depends also on the joining time and the joining force. The joint stability cannot be concluded from the width of the diffusion zone and has yet to be proved in stability tests.

The microsection of the diffusion joined dissimilar material joint, shown in Fig. 8, shows the indents in the joining materials and in the joining zone. The Vickers hardness HV 0.1 of the dissimilar material joint was determined with the joining parameters T = 500[degrees]C, t= 12 h by means of a small load hardness measuring device Zwick in compliance with EN ISO 6507-1.

[FIGURE 6 OMITTED]

[FIGURE 7 OMITTED]

[FIGURE 8 OMITTED]

In diffusion joining the material is stressed through a thermal process. This can either lead to an increase or decrease of the hardness, which affects the joint strength. Prior to the experiments the initial hardnesses were determined (CrNi steel 178 HV 10, AlMg alloy 60 HV 10) on thermally non-stressed joining partners. Figure 9 shows the hardness progression of the dissimilar material joint after the diffusion joining.

The examinations showed that the hardness in the joining materials is nearly constant. The hardness of the CrNi steel increases sharply near the joining zone up to 250 HV0.1. The Vickers hardness forms a step in the joining zone. This hardness increase is a sign of intermetallic phases in the joining zone. The thickness of these phases strongly depends on the joining time (Fig. 10). The intermetallic phase is about 20 [micro]m thick. On the steel side it is mainly [Fe.sub.2][A1.sub.5] and on the aluminium side Fe[Al.sub.3]. The intermetallic phase Fe[Al.sub.3] forms above 350[degrees]C, has an aluminium concentration between 74-76 atom% and a complex monoclinic, bodycentred cubic structure. Furthermore, this phase is characterized by a columnar crystal shape. In contrast, the aluminium concentration in the intermetallic phase [Fe.sub.2][Al.sub.5] lies between 69-73 atom%. For comparison, the aluminium concentration in the joining zone of the sample, which was joined at 500[degrees]C, was ca 59 atom%.

[FIGURE 9 OMITTED]

[FIGURE 10 OMITTED]

[FIGURE 11 OMITTED]

Figure 11 shows a broken joint after a stability test. An oxide layer, based on aluminium oxide, could be detected on the surfaces. The residual oxygen concentration in the pre-vacuum of 10-1 mbar caused the creation of the oxide layer, which was not broken due to the low joining temperatures [5]. These oxide layers additionally function as diffusion barriers and extend the joining time.

4. CONCLUSIONS

The examinations proved that through diffusion joining interlayer-free and crack-free dissimilar material joints can be manufactured between CrNi steel (X5CrNi18-10) and AlMg alloy ([AlMg.sub.3]) at joining temperatures lower than 600[degrees]C. Ground and polished surfaces are a precondition for an optimal joining process.

Different hardness and expansion behaviour (different expansion coefficients) of the joining parts enabled the manufacture of homogeneous and mechanically stable joints. Hardness and expansion behaviour are closely linked with each other and they are essential for achieving a low-tension joint. The creation and the characteristics of the joining zone depend on the Mg and Al content as well as on the alloy elements of the CrNi steel. The joining parameters (temperature, time and joining force) show functional interdependence, which results in the creation of different diffusion areas. The process parameters have a direct influence on the creation of intermetallic phases in the joining zone. Currently no conclusions can be made on the creation of intermetallic phases in a system with a variety of elements like in the examined case. That is why the examinations on the creation of intermetallic phases were restricted mainly on the system Fe-Al. The creation of intermetallic phases in the Fe-Al system is based on the almost absolute insolubility of iron and aluminium in the solid state. A short heat impact stimulates mainly the formation of intermetallic phases [Fe.sub.2][Al.sub.5] and Fe[Al.sub.3].

Examinations on intermetallic Al-Fe phases as well as the inclusion of further alloy elements and the prevention of an oxide layer on the Al[Mg.sub.3] will be subject of future research.

ACKNOWLEDGEMENTS

The described results are extracts from an ongoing research project within the frame of the BMWi programme INNO-WATT--FuE-Projekt Reg. VF080016 "Diffusion joining of ceramics" and FuE-Projekt Reg. IWO91046 "Verfahrensentwicklung zum Diffusionsschweissen von Werkstoffen mit unterschiedlichen Eigenschaftsprofilen", Germany.

doi: 10.3176/eng.2010.4.03

REFERENCES

[1.] Pohle, C. Schweissen von Werkstoffkombinationen. DVS-Verlag, Dusseldorf, 1999.

[2.] Broda, T., Herrmann, J. and Uhlemann, M. Fugetechnologien fur Mischverbindungen durch Lotapplikation mittels Ultraschallschweissen. DVS-Berichte, Band 2009 "Die Verbindungs-spezialisten Tagungsband". Verlag DVS Media GmbH, 206-212.

[3.] Wilden, J. and Bergmann, J.-P. Herstellung von Titan-Aluminium and Titan-Stahl-Verbindungen durch Diffusionsschweissen. Mitteilung aus der Technischen Universitat Ilmenau, Fachgebiet Fertigungstechnik, Schweien and Schneiden, 2004, 56, 199-207.

[4.] Dahms, S., Gemse, F., Basler, U., Martin, H.-P. and Triebert, A. Diffusion joining of silicon nitride ceramics. Estonian J. Eng., 2009, 15, 301-308.

[5.] Dahms, S., Gemse, F. and Basle-, U. Diffusionsschweissen--ein innovatives Fugeverfahren fur Werkstoffverbunde mit unterschiedlichen Eigenschaftsprofilen. DVS-Berichte, Band 2009 "Die Verbindungsspezialisten Tagungsband". Verlag DVS Media GmbH, 54-59.

[6.] Pietrzak, K., Kalinski, D. and Chmielewski, M. Interlayer of A1203-Cr functionally graded material for reduction of thermal stresses in alumina-heat resisting steel joints. J. Europ. Ceram. Soc., 2007, 27, 1281-1286.

[7.] Travessa, D., Ferrante, M. and den Ouden, G. Diffusion bonding of aluminium oxide to stainless steel using stress relief interlayers. In Materials Science and Engineering. Elsevier Science B.V., 2002, A337,287-296.

[8.] Zeng, L., Case, E. D. and Crimp, M. A. The interfacial microstructure of joined single crystal and polycrystalline alumina. In Materials Science and Engineering. Elsevier Science B.V., 2003, A360,228-236.

[9.] Geguzin, J. E. Grundzuge der Diffusion in Kristallen. VEB Deutscher Verlag fur Grundstoff-industrie, Leipzig, 1977.

Steffen Dahms (a), Priit Kulu (b), Renno Veinthal (b), and Ursula Basler (a)

(a) Gunter-Kohler-Institut fur Fugetechnik and Werkstoffprufung GmbH, Otto-Schott Str. 13, 07745 Jena, Germany; {sdahms, ubasler}@ifw-jena.de

(b) Tallinn University of Technology, Department of Materials Engineering, Ehitajate tee 5, 19086 Tallinn, Estonia; {priit.kulu, renno.veinthal}@ttu.ee

Received 26 April 2010, in revised form 17 September 2010

Table 1. Thermophysical properties

Parameter                           CrNi steel          AlMg alloy
                                   X5CrNi18-10            AlMg3

Density, g/[cm.sup.3]            7.93               2.7
Space lattice                    fcc                fcc
Melting point, [degrees]C        1400-1455          620
Thermal expansion at             18 x [10.sup.-6]   23.5 x [10.sup.-6]
  20-100[degrees]C, [K.sup.-1]
Heat conductivity at RT,         16.3 at RT         125
  [Wm.sup.-1] [K.sup.-1]
Vickers hardness, HV             175                75
Modulus of Elasticity, GPa       190-210            71
Tensile strength, MPa            460-1100           300-450

Table 2. Experimental joining parameters of the dissimilar
material joint CrNi steel with AlMg alloy

No. of the   Joining temperature   Joining time t,   Joining force F,
experiment      T, [degrees]C             h                 N

1                    450                  2                2000
2                    500                  2                2000
3                    550                  2                2000
4                    450                 12                1000
5                    500                 12                1000
6                    550                 12                1000

Table 3. Surface characterization of the dissimilar
material joint of CrNi steel with AlMg alloy

Surface               CrNi steel
parameters
                      Manufacturing quality   Final state

[R.sub.a], [micro]m           0.834              0.109

Surface               AlMg alloy
parameters
                      Manufacturing quality   Final state

[R.sub.a], [micro]m           0.298              0.225