When nickel brazing Inconel 718 base metals in a vacuum furnace some of the difficulties experienced with the brazing include base metals that come out of the furnace dark and discolored, and the brazing filler metal doesn't wet the surfaces well. In this article we will explain why this is happening, and what can be done about it?
T-specimens: The specimen on top is typical of the way an Inconel 718 discolors and the BFM does not flow out.
This is not an uncommon problem with a variety of base metals containing small amounts of titanium and/or aluminum. Both titanium and aluminum will easily oxidize, and once those oxides are formed they cannot be easily removed in a standard vacuum-furnace atmosphere. Yes, vacuum is an "atmosphere" in normal production environments since the level of vacuum in the furnace during typical brazing is such that there is, relatively speaking, a goodly number of air molecules still in the furnace, including moisture in that air. Of course, moisture represents the presence of oxygen, which can indeed react with either titanium or aluminum to form very tenacious titanium oxides and aluminum oxides on the surface of the base metal, which will inhibit or prevent brazing filler metal (BFM) flow.
Traditionally, we get concerned whenever the percent of titanium and/or aluminum is about 0.4% or greater in any base metal that is to be brazed. Inconel 718 contains approximately 0.9% titanium and 0.6% aluminum.
There are several ways to deal with these small amounts of titanium and/or aluminum so that they will not interfere with the brazeability of the base metals in normal vacuum brazing furnaces. Two common ways are nickel plating and fluoride-ion cleaning. Additional methods such as nickel-alloy grit blasting are sometimes used, but I want to concentrate on the two I've mentioned above for now. In this week's blog let's examine the first method, namely, the use of nickel plating to prepare surfaces for brazing.
Nickel plating is one of the most common methods to prepare certain base-metal surfaces for brazing. Nickel does not readily oxidize. Therefore, any nickel-plated surface will generally remain oxide-free, and the applied BFM can alloy nicely with the nickel plating forming a good metallurgical bond.
First of all, please be aware that there are two types of nickel-plating that can be used in preparing surfaces for brazing – electrolytic and electroless.
Put simply, electrolytic nickel plating is pure nickel. Pure nickel has a melting point of approximately 2600°F (1300°C). Therefore, electrolytic nickel platings are a preferred choice when brazing temperatures will be above 1700°F (925°C), which is typical for many aerospace applications.
Electroless nickel, however, is actually an alloy of nickel and either phosphorus (the most common) or boron. Thus, such plated coatings have a melting point in the vicinity of only 1600°F (870°C), depending on how much phosphorus or boron is in the electroless nickel being used for the plating. Electroless-nickel plating should only be used when brazing temperatures will be low – 1200-1500°F (650-800°C) – so that the plating itself will not melt and flow away.
When using either type of nickel plating, it may not be necessary to plate the entire part, especially if the parts are very large. Instead, the plating can be limited to just the faying surfaces (the surfaces that will be bonded together inside the brazed joint).
The thickness of nickel plating applied is an important consideration, and the thickness that should be specified will vary according to length of time of the brazing cycle and the temperature that will be used in the furnace. The higher the temperature, or the longer the brazing cycle, the thicker the nickel plating should be because the base-metal constituents underneath the nickel plating will diffuse into the plating and attempt to reach the outside surface, where they can once again react with the furnace atmosphere to form their desired oxides of titanium and/or aluminum.
The brazer must be sure that the plating is thick enough so that the diffusion of titanium or aluminum will NOT get through that plating layer before the brazing cycle is completed. Therefore, depending on the length of the brazing cycle and the temperature at which the brazing occurs, plating thickness requirements may vary from about 0.0002-0.0015 inch (.005-.04mm). This is further compounded by the fact that the nickel plating is also diffusing into the base metals! So, the actual amount of nickel plating may have to be determined by experimentation.
To verify the "goodness" of a nickel plating (primarily relates to electrolytic nickel), it is often wise to run a "blister test" on the plating before you commit the plated parts to production in your brazing shop. The purpose of this test is to verify how adherent the plating is to the base metal. Plated assemblies (or test pieces having similar cross sections as the major components being plated) are tested at elevated temperatures in your brazing furnace (usually above 1000°F/550°C) to see if any blisters appear on the surface of the plating. Such blisters occur because of the outgassing of contaminants on the surface of the base metals underneath the plating that were not removed from metal surfaces by cleaning prior to nickel plating.
A good, reliable plater should always strive to be sure that the surfaces to be plated have been thoroughly cleaned to remove anything that could impede the bonding of the nickel-plating to that surface. Otherwise, during the brazing process, the BFM may melt and alloy nicely to the plating. But if the plating lifts off the base metal because of blistering, the joint may fail in service!
Now let’s examine the topic of fluoride-ion cleaning of certain nickel-based superalloys to remove titanium and/or aluminum from the surface of the base metals prior to brazing.
Traditionally, as mentioned above, we get concerned whenever the percent of titanium and/or aluminum is about 0.4% or greater in any base metal that is to be brazed. Inconel 718 contains approximately 0.9% titanium and 0.6% aluminum.
To handle this concern in brazing, I mentioned two good ways to deal with this. One is by nickel plating the Inconel surfaces that will be brazed, and the second method involves the use of fluoride-ion cleaning procedures to achieve a readily brazeable surface. Let's look at this in more detail.
Surface morphology related to oxides layers that cause brazing problems. Images of polished Inconel 718 substrate (a), Ti oxide coated (b), and Al oxide coated (c) Inconel 718 surfaces
Also known as "FIC," "FIC cleaning" (a double use of the word "cleaning") or simply as "F-Clean," this is a relative newcomer to the world of cleaning processes for braze preparation, having really come into its own in the late 1980s. The process utilizes HF (anhydrous hydrogen fluoride gas) to effectively remove Cr, Ti and Al from the surfaces of nickel superalloy materials down to a depth typically ranging from 0.0005-0.0015 inch (0.015-0.05mm), i.e. about 8-30 microns.
Depletion of the Cr, Ti and Al is necessary so that the readily formed oxides of these metals cannot interfere with the wetability of the base metals by the brazing filler metals involved in the joining process.
The FIC process is performed in a special retort chamber that looks very much like a typical vertical-loading vacuum-furnace setup. Actual operating temperatures for this process are about 2000°F (1100°C), and the pressure in the chamber can vary from about 100 Torr (about 1/8 atmosphere) up to full atmospheric pressure.
This cleaning process has become very popular for use with braze repair of gas-turbine components for both aerospace and ground-power turbine applications and represents, in my opinion, the current state-of-the-art in cleaning surfaces and deep cracks in most superalloys today. As a repair technique, it is highly effective at removing oxides and other foreign matter from deep within cracks on the surface of these metals. Thus, it is an excellent method to "deep clean" surfaces of aero components coming back from service for overhaul and repair. The nice thing about this process is that, when used properly, it does not attack nickel or cobalt and, thus, does not have a negative impact on the dimensional control of such surfaces.
Metal components coming out of the FIC furnaces are very bright and shiny, and I have personally never seen metal look so shiny, bright, clean, etc. as metal parts do when exiting from this process. And, since the surfaces are depleted of Cr, Ti and Al, they can remain in this cleaned state at room temperature for long periods of time with no significant change in surface coloration or cleanliness since the surface of such FIC superalloys is essentially pure nickel and/or cobalt. Thus, they behave almost as if they had been nickel plated!
To View a listing of all Dan’s articles please click here
Read Dan Kay's Biography
© Copyright Dan Kay 2010