I. Specification of the problem
The present Project is dealing with the fabrication of electrical (high-voltage) feedthrough consisting of YSZ (ZrO2 + 3 % mole Y2O3) body (insert) with fifteen (15) Pt-Pd pins (contacts) positioned inside the stainless steel (AISI 316L) holder. In Fig. 1, drawings of the electrical feedthrough assembly (“first version”) together with the constituent components are given.
Fig. 1: Constituents of the electrical feedthrough assembly in (a), (b) and (c) together with a side-view cross-section (d).
In fact, the problem is reduced to the joining of the Zirconia-based ceramic insert (with fifteen (15) Pt-Ph contacts) to the AISI 316L holder along the mating surfaces as indicated in Fig. 2. The product should be able to operate at 250 ºC.
Fig. 2: An example of electrical feedthrough assembly (a) together with a side-view cross-section (b) and zoomed area showing the location for placing brazing (solder) filler.
Before proceeding with the discussion, it seems worthwhile to provide some basic information about materials are to be joined.
1.1 Austenitic stainless steel grade AISI 316L
The AISI 316L is an austenitic stainless steel is widely used in applications that require a degree of resistance to crevice and/or pitting corrosion. The “L” identifier of 316L indicates lower Carbon content than in the standard AISI 316 grade, a characteristic which reduces the susceptibility to sensitization (i.e., grain boundary carbide precipitation during exposure to an intermediate temperature somewhere between 550 and 800 ºC [1]).
Typical composition of 316L steel is given in Table 1. Note. The presence of Molybdenum makes steel more resistant (although more expensive) to the corrosion in the operational environment than, for example, AISI 304 grade.
Table 1: Chemical composition (%) of grade AISI 316L stainless steel
Corrosion resistance of stainless steel is a result of the presence of a thin oxide layer on its surface. The passivation of stainless steel takes place under atmospheric conditions, which yields an oxide film that is self-healing on localized damage. The oxide, naturally formed in the atmosphere, is generally referred to as the native oxide and it is affected by environmental factors. It is also relevant to mentioned here, that thermal expansion of the AISI 316L steel is ~15.9.
1.2 Zirconia-based ceramics
Much of the recent technological interest in Zirconia-based ceramics stems from the fact that ZrO2 (pure) can exist in three (3) different crystal structures (allotropic modifications): cubic, tetragonal and monoclinic (Fig. 3).
Fig. 3: Zirconia allotropes and the phase transition temperatures [2].
The structure (or combination of structures) that exists in ZrO2-based ceramics can be controlled by alloying and heat-treatment to produce materials with high strength and relatively high toughness.
Zirconia ceramics have found broad applications in a variety of energy and biomedical applications because of their unusual combination of strength, fracture toughness, ionic conductivity, and low thermal conductivity. These attractive characteristics are largely associated with the stabilization of the tetragonal and cubic phases through alloying with aliovalent ions. The large concentration of vacancies introduced to charge compensate of the aliovalent alloying is responsible for both the exceptionally high ionic conductivity and the unusually low, and temperature independent, thermal conductivity.
The high fracture toughness exhibited by many of zirconia ceramics is attributed to the constraint of the tetragonal-to-monoclinic phase transformation and its release during crack propagation, i.e., to the stress-induced phase changes that can occur between the different crystal structures.
Excellent resistance to thermal shock and mechanical impact damage has led to the use of PSZ (partially stabilized zirconia) materials as extrusion die materials. Superior wear properties and low susceptibility of stress corrosion has led to the use of PSZ as a structural biomedical material. Wear properties, high strength, low thermal conductivity (2.0 – 3.3 W/m.K) and relatively high thermal expansion (CTE ~ 9.6 x 10-6 K-1) have resulted in the application of PSZ in a variety of internal combustion engine components.
It is important to mentioned here that structure transformations are accompanied by volume changes which may cause cracking if cooling/heating is rapid and non-uniform. Additions of some oxides (MgO, CaO, Y2O3) to pure zirconia depress allotropic transformations (crystal structure changes) and allow to stabilize either cubic or tetragonal structure of the material at any temperature.
Depending on sintering temperature and other processing parameters, the following forms of stabilized zirconia may be prepared:
- Fully stabilized zirconia (FSZ) with cubic crystal structure
- Partially stabilized zirconia (PSZ) with mixed structure (cubic + tetragonal)
- Polycrystalline tetragonal zirconia (TZP) with metastable tetragonal structure of very fine zirconia grains sintered at low temperature
- Composites of zirconia with other ceramics such as Al2O3 containing ZrO2-particles, a material which is referred to as zirconia-toughened Al2O3 or ZTA.
The most popular stabilizing addition to zirconia is yttria (Y2O3), which is added and uniformly distributed (up to 5.15 wt. %). Principles of the “stabilization” are explained schematically in Fig. 4 and quasi-binary ZrO2 -Y2O3 phase diagram is shown in Fig. 5 [3].
Fig. 4: Effect of Y2O3 on “stability” of cubic (fluorite-type) structure in the ZrO2 -Y2O3 system (schematically).
Fig. 5: The quasi-binary ZrO2 – Y2O3 phase diagram [3]. Note. The letters C, T and M on the diagram are stated for cubic, tetragonal and monoclinic structure, respectivel
As mentioned above, in the present project, partially stabilized zirconia (PSZ) containing 3 mole % of Y2O3, (i.e., this ceramics has a mixed structure), is used as a material of the ceramic body (Fig. 1).
II. Active brazing of ZrO2-based ceramics to AISI 316L stainless steel
2.1 Active brazing of Zirconia
It is generally accepted that joining of Zirconia-based ceramics can be accomplished by (active) brazing using Ti-activated Ag-Cu based alloys [4]. Titanium is the “activator” used in commercial brazes, reflecting the singular characteristic that it can form very wettable reaction products. The general use of Ag-Cu alloys rather than pure Copper as a solvent is understandable in terms of effects on the solubility and thermodynamic activity of Titanium and hence, the ease of forming a wettable reaction product. Similarly, the introduction of Indium and Tin as quaternary alloying elements is consistent with their greater enhancement of Titanium activity, and it was identified TiO as the reaction product formed by the Ag Cu26 Sn4 Ti4 and Ag Cu23 In15 Ti1.5 alloy with ZrO2.
The Ag-Cu solvents also offer other advantages. The preferred composition Ag Cu28 is a eutectic, and hence, has a “singular melting temperature” of 780 ºC. The solvent alloy is ductile, enabling pre-forms to be fabricated readily and conferring on joints some ability to accommodate thermal mismatch stresses. The Titanium-additions can affect the alloy liquidus temperature, causing the liquidus and solidus to “separate”, so that melting involves a transitional two-phase (liquid + solid) stage and stiffening and hardening the solidified filler alloy. The introduction of Tin and Indium will also stiffen and harden the solvent, but the Ag Cu23 In15 Ti1.5 alloy is described as being ductile, so that these complex alloys can still have attractive or, at least, acceptable properties.
2.2 Brazing of AISI 316L stainless steel using Ti-activated Ag-Cu based filler alloy
It was demonstrated that austenitic stainless steels can be brazed using Ti-activated filler metal, like for example, Ag Cu27.25 In12.5 Ti1.75 alloy [5]. This information seems to be confirmed by the results of the (preliminary) experiments conducted at Mat-Tech BV in the framework of this Project.
In Fig. 6 a series of bright-field optical images of the brazing seam in the YSZ / AISI 316L joint fabricated in vacuum using Ti-activated filler alloy Ag Cu35.25 Ti1.75, with liquidus and solidus temperature 815 ºC and 780 ºC, respectively.
It turned out the product brazement is not leak (vacuum) tight, which is not very surprising, given the fact that during the brazing procedure, gap between the mating ceramic and steel surfaces was insufficiently filled (Figs. 6a and 6b). However, in the areas of the brazing gap where contact between the molten filler and the mating surfaces was established, the active (reaction) bonding was successful as can be appreciated by looking at Figs. 6c and 6d.
The attempt to improve quality of the joint by increasing peak brazing temperature to 910 ºC was in vain. The brazement fabricated according to this schedule was also not leak tight.
Fig. 6: Bright-field optical images of the brazing seam developed during fabrication of the joint between YSZ and AISI 316L stainless steel in vacuum using Ag Cu35.25 Ti1.75 filler alloy: a) and b) general views of the brazement cross-section showing insufficient filling of the brazing gap by the filler metal and c) and d) magnified areas of the brazing seam confirming sound bonding in some areas of the product joint.
Another issue that might cause uneasiness in the manufacturing of the feedthrough assembly using active brazing technique is the surface discoloration of the stainless steel part, i.e., appearance of heat-tint colors upon exposure to the furnace atmosphere at elevated temperature during the brazing cycle. This phenomenon affects not only an aesthetic appearance of the brazement, but, as it turned out, may have (according to the Client) a serious implication for the subsequent micro-welding of the (surface tinted) AISI 316L component.
III. On the possibility of soldering of YSZ ceramic insert to AISI 316L stainless steel holder
3.1 General considerations and modification of the AISI 316 steel surface prior to soldering
Obviously, the use of much lower joining temperature, which is pertain to the soldering process may obviate the issues connected with the stainless steel holder discoloration. Given very high (for conventional solder interconnects) service temperature of the product joint, viz. 250 ºC, the only option left in the selection of suitable filler alloys is to use for joining Zirconia (YSZ) insert to AISI 316L holder an eutectic (280 ºC) AuSn20 solder.
However, one caveat is in order here. Since the product feedthrough must be able to operate in a very high vacuum, the use of flux during the soldering is highly undesirable. In other words, the
“fluxless” soldering process must be developed. To this end, we suggest conducting joining (soldering) in an inert (protective gas) atmosphere or in vacuum at 330-340 ºC using AuSn20 filler metal, with the mating surfaces of the ceramic insert as well as stainless steel holder are to be Au-plated (finished).
In the case, when the presence of Nickel within the electrical feedthrough assembly is allowed, the surface modification of the AISI 316 stainless steel part can be accomplished by the Ni-deposition (perhaps, electroless Ni(P)) with the subsequent Au-finish. If, on the other hand, the use of Ni is prohibited, then matter becomes rather complicated. This issue should be discussed with the Client.
3.2 Metallization of the ZrO2-based ceramic component
The standard, a so-called “Mo/Mn process” developed for metallization of various oxide ceramics cannot be used for the purpose of the present Project. The main objection here is that it is not clear what will happen at the contact surface between ZrO2 and Pt-Rh alloy (pins) during high-temperature exposure (up to 1300 ºC) to the “wet” hydrogen atmosphere. Note. The latter is an essential step in the Mo/Mn process. It is possible that even in a wet hydrogen, interaction between ZrO2 and Pt-Rh alloy may result in the formation of intermetallic phases, which will have a deleterious effect on structural integrity of the assembly.
Perhaps, a so-called “thick-films” metallization can be a viable solution. A thick-film (“hybrid”) metallization is constructed on a ceramic substrate by using fine metallic (e.g., Ag, Ag/Pd) particles mixed with binding materials (low-melting point glass powder, a so-called frit) to promote adhesion to the substrate and to hold the metal particles in contact [6]. The metal particles and binding material are mixed with an organic material to form a relatively thick mixture called paste or ink. The majority of the commercially available pastes are thixotropic, i.e., these pastes are thick (viscous) under normal conditions, but flow (become thin, less viscous) over time when shaken, agitated, or otherwise stressed.
Generally, the binding material consists of fine particles with a low melting point, such as lead-borosilicate glass (“solder glass”). Most of the solder glasses have a lead oxide (PbO) content ranging from 70 to 85 wt.%, boron oxide (B2O3) from 10 to 20 wt.%, and silicon dioxide from 10 to 20 wt.%. Thermal expansion may range from 8 to 12 × 10-6 K-1, and “softening point” is between 350 and 500 ºC. It is also relevant to mention that a small addition of ZnO and Al2O3 is frequently used to modify the desired properties and to improve the chemical stability of the glasses. The glass holds the metallic particles in contact and promotes adhesion to the ceramic by reacting with the substrate during the firing
process (usually at ~850 -900 ºC). The glass (frit) has a lower “melting” point than the metallic particles (Ag or Ag/Pd) and, during firing, wets the metallic particles, suspending them in contact with each other. Physico-chemical interaction is also taking place between the glass and the ceramic substrate.
After application of, for instance, Ag-based paste on the ceramic surface and drying in air at 150 ºC for 10-30 min, the “green” metallized substrate is fired in air at peak temperature 850 ºC for 10 min. More details can be found in the specifications provided by the paste suppliers, e.g. [7].
During the firing process, the glass flow occurs from the surface layer (paste containing glass frit) to the ceramic, filling the space (capillary) between the metallization layer and the substrate surface as well as the internal pores present in the layer. At the same time, sintering of the metal particles also takes place. The resultant structure after firing is shown schematically in Fig. 7: the metal particles are bound together and to the substrate by the glassy phase, and this is particularly important at the substrate-paste interface.
Fig. 7: Schematic structure of a fired “thick-film” metallization.
However, it is important to realize that the described above technology has been developed for metallization of Alumina- and Beryllia-, but not Zirconia-substrates. As to the application of the conventional “thick-film” method for metallization of ZrO2-based ceramics, it seems that this technique can also be used in the case at hand as well no matter whether PbO-containing or “Pb-free” frit based on Bi2O3 is used. This conclusion is based on the phase diagram information on quasi-binary PbO – ZrO2 and Bi2O3 – ZrO2 systems available in the literature (Fig. 8).
Fig. 8: The quasi-binary phase diagrams: a) PbO – ZrO2 [8] and b) Bi2O3 – ZrO2 [9]
One can see that a liquid phase that provides adhesion to the Zirconia-substrate is formed in both systems in the temperature range pertain to the firing procedure, i.e., at 800 – 900 ºC (as suggested by the paste manufacturers). However, it is to be mentioned that fabrication of the “thick-film” metallization on cylindrical surface is by no mean a simple and straightforward procedure for any ceramic material. This might be a serious issue.
Further, a layer of Nickel (if, the use of Ni is not prohibited) is deposited on the top of the “thick-film” Ag- or Ag/Pd- based metallization followed by the Au-plating (finish). Then, soldering of Au-finished components of the electrical feedthrough assembly can be accomplished in vacuum (or in a protective atmosphere with low oxygen potential) at 330-350 ºC using AuSn20 solder. It is likely that under these conditions, discoloration of the AISI 316L surface can be avoided.
3.3 On the idea of using solder glasses
This idea was “flying in the air” for some time. However, it is important to stress here that in this case, the soldering process is to be carried out in air at the temperature range 410 – 700 ºC [10]. When stainless steel is heated to high temperatures, the native oxide layer can thicken, causing a so-called “rainbow tint”. Fig. 9 below represents the temper (heat-tint) colors that are likely to form on stainless steel type 1.4301 (AISI 304) if heated in air. This information, however, must be treated with
care when interpreting the heat tint colors observed on stainless steel surfaces as the heating conditions are not specified.
Fig. 9: Oxidation colors (heat tints) formed on AISI 304 stainless steel heated in air.
Clearly, soldering with solder glasses will result in discoloration of stainless steel surface. (Note. It is important to keep in mind that in the present project we are dealing with different grade of austenitic stainless steel, namely AISI 316L. However, the discoloration behaviour of this stainless in air is not expected to be much different from that described for the AISI 304 steel.)
Another question that might cause uneasiness is that it is established that thermal expansion coefficient of the solder glass must necessarily be matched to the expansion coefficient of the components to be joined. But it is generally chosen to lie somewhat below the expansion coefficient of the sealing partners. As a rule, the coefficient of thermal expansion of the solder glasses should be by Δα = 0.5-1.0 ×10-6 K-1 smaller than the expansion coefficient of the sealing partners, which is impossible to accomplished, given the relatively large difference between thermal expansion of ZrO2 (CTE ~ 9.6) and stainless steel AISI 316L (CTE ~15.9).
IV. Final Thoughts
It appeared that active brazing using Titanium-activated Ag-Cu-based filler alloys is certainly a viable option for joining of Zirconia (YSZ) inserts to AISI 316L stainless steel holder, since the product electrical feedthrough assembly should be able to operate at 250 ºC. The issues connected with the placing of the filler metal into the brazing clearance can, in principle, be solved by using a somewhat thicker brazing foil.
As to the discoloration of the AISI 316L stainless steel surface during the brazing process, it turned out that there exists some experience at Mat-Tech BV as to the appearance of heat-tint colors upon exposure to the furnace atmosphere [11].
References
[1] D. A. Jones, Principles and Prevention of Corrosion, Macmillan, Inc., 1992
[2] W. Borchardt-Ott, Crystallography, 2nd Ed., Springer, 1995
[3] J. Chevalier, L. Gremillard, A. V. Virkar, D. R. Clarke, The Tetragonal-Monoclinic Transformation in Zirconia: Lessons Learned and Future Trends, J. Am. Ceram. Soc., 92 (2009) 1901-1920
[4] M. G. Nicholas, Joining of Ceramics, The Institute of Ceramics, Chapman and Hull, 1990
[5] A. Guedes, A.-M. Pires Pinto, Active Metal Brazing of machinable Aluminium Nitride-Based Ceramic to Stainless Steel, J. Mater. Eng. Perform., 21 (2012) 671-677
[6] R. W. Vest, Material science of thick film technology, Ceram. Bull., 65 (1986) 631-636
[7] www.heraeus.com
[8] B.-K. Koo, P. Liang, H.-J. Seifert, F. Aldinger, Thermodynamic Assessment of the PbO-ZrO2 System, The Korean Journal of Ceramics, 5 (1999) 205-210
[9] T. Takamori, M. W. Shafer, J. Am. Ceram. Soc., 73 (1990) 1453-1455
[10] www.schoot.com
[11] M. Biglari, private communication, May, 24, 2022
