Fig. 1 – Note that small, interstitial atoms cannot substitute for the larger atoms in the matrix of atoms shown above, but must fit into the small intersticies (spaces) between those larger atoms, as shown in this idealized conceptual diagram of atoms lining up with each other.

Overview: During vacuum brazing with nickel-based brazing filler metals (BFMs), it is possible to hold the brazed parts at brazing temperature long enough for the BFM to solidify completely while being held at brazing temperature! The key is “diffusion”, and involves tiny interstitial atoms in the BFM.

Isothermal solidification can be a very useful brazing process for some brazing filler metals (BFMs), and can result in a significant increase in the re-melt temperature of the BFM in that brazed joint.

To better understand the process, let’s first examine the component parts of the phrase “isothermal solidification”. “Iso” essentially means “ equal, or the same”, and “thermal” of course refers to temperature. So we’re looking at a BFM solidification process in which that solidification takes place while the furnace is being held at the same, steady temperature! Although that may sound strange, there’s some real logic to it. Isothermal solidification (we’ll refer to it as ITS in this article) depends a lot on the diffusion capabilities of various components of the BFM while that BFM is being held at the brazing temperature.

Nickel-brazing of aerospace components in a vacuum furnace is an excellent example of how this process might prove useful for brazed-components that could be subjected to sudden, accidental high temperature extremes (much higher than the original brazing temperature) during end-use service. The AWS A5.8, Class BNi-2 (also known in the industry as AMS 4777) is an example of a Ni-based BFM whose re-melt temperature can be significantly increased by isothermal solidification (ITS). The amount of the increase could, as an example, range from 200F-700F (100C-400C) above the actual brazing temperature originally used to join the parts! This particular BFM chemistry has almost 3.5% of boron (B) added to it as a temperature-depressant, i.e., as an element added to the BFM to significantly lower the initial melting point of the BFM chemistry (it does this by forming low-melting eutectic compositions with the nickel into which it is alloyed). Boron is a very small atom compared to the much larger atoms of nickel, chromium, iron, and silicon that make up the rest of the BNi-2 chemistry. Thus the boron atom does not fit as a so-called substitutional-atom in the matrix of the BNi-2 alloy, but is, instead, known as an interstitial atom, i.e., one that fits into the small spaces between the much larger atoms, as shown in Figure 1.

Interstitial atoms are not as strongly bonded into the matrix of atoms as substitutional-atoms, which means that “interstitials” can enter or leave the matrix of atoms much more easily than substitutional atoms can. As you know, as a metal gets hotter and hotter, it will expand. It does this because atoms vibrate in place more and more when heated, occupying more and more space to do so. Thus the overall dimensions of the metal get larger and larger as the metal gets hotter and hotter, since the spacing between each of the atoms in the alloy is increasing.

Simplified metallurgical phase diagram of the nickel-boron system.

Fig. 2 – Simplified metallurgical phase diagram of the nickel-boron system.

Here’s where the tiny size of the boron atom comes into play! Because it is so small, and only weakly “bonded” into the BFM alloy structure (because it is an interstitial atom), the boron atom is able to escape (i.e., diffuse away) from the BFM when the brazing temperature has increased to the point where the spacing between the larger substitutional atoms is great enough for the small boron atoms to get through!

Remember, the boron was added into the BFM to lower its melting point. Therefore, does it not seem logical that when the boron leaves (i.e., diffuses away from) the BFM, the melting point of the BFM should go back up? In fact, that does actually happen.

To understand this a little bit better, let’s get into some of the metallurgy of brazed joints. Let’s look briefly at a so-called phase-diagram of the Nickel-Boron alloy system as shown in Figure 2.

The varying amount of boron in the alloy is shown along the bottom of the graph, and temperature is shown on the vertical axis. It can be seen that as the amount of boron that is added to the alloy increases from zero up to 3.5%, the curved line labeled “liquidus” drops significantly, from about 2650F (1455C) down to only 2000F (1093C), whereas the line labeled “solidus” remains pretty steady from left to right at 2000F (1093C). Let’s look at the same diagram in Fig. 3, with the vertical lines A (2% B), and B (3.5% B) added to it, as well as a horizontal line representing a theoretical brazing temperature.

At 2% boron content, you can see that as you increase the temperature of that specific alloy-composition from room temperature up to 2000F (1093C), you reach the solidus-temperature line. The solidus temperature is simply the temperature below which the alloy remains completely solid (hence the word “solid”-us). Thus, as soon as you cross the solidus temperature line during heating, the BFM will start to melt! It will continue to melt further and further until you reach the curved line labeled “liquidus”, which, as you guessed it, is the line representing the temperature, above which, that alloy chemistry is supposed to be fully liquid (hence the word “liquid”-us). Notice I said “supposed to be”. Please see my article about liquation in the AWS Welding Journal (Sept., 2010), and also in the blog-article I wrote for the VacAero website entitled Liquation: Good or Bad? for a more thorough discussion on this interesting topic. Technically speaking, the liquidus temperatures are usually determined by cooling a liquid BFM and determining the temperature at which a particular composition begins to solidify.

Ni-B phase diagram with a hypothetical brazing temp shown.

Fig. 3 – Ni-B phase diagram with a hypothetical brazing temp shown.

Now, back to our discussion of ITS. Notice the difference between vertical lines A and B in fig. 3. On vertical line B the liquidus and solidus temperatures are the same. Such a junction is called a “eutectic point”, and represents the lowest melting point for a given BFM alloy system, i.e., it is the composition at which the lowest melting temperature BFM-liquid can exist.

Notice now the horizontal dotted-line in Fig. 3 that is labeled “Brazing temp”. The brazing temperature used in any brazing operation should be at least 100F (or 50C) higher than the “melting point” (liquidus temp.) of the BFM being used. We’ll assume that’s the case in our present example.

Now, remember that at brazing temperature the boron begins to diffuse away from the joint into the base metals of the part being brazed (as the boron atoms move, i.e., diffuse away, to achieve an equilibrium balance of boron throughout the entire structure). As you can see, the liquidus line of the BFM in Fig. 3 begins to rise as you decrease the amount of boron in the joint from just above 3.5% down to 2% or less, i.e., as the boron continues to diffuse away from the braze-joint area. The boron concentration in the brazed joint won’t go to zero, since the boron (B) atoms are merely trying to achieve an equilibrium balance throughout the structure.

Notice in Fig. 3 that the horizontal dotted line representing our hypothetical brazing temp crosses the liquidus line at about 2.5% boron (vertical line C). When the brazing cycle is held at brazing temp long enough until the boron in the BNi-2 BFM has diffused away to below 2.5%, there is then not enough boron left in the joint to keep the BFM liquid at that specific brazing temp. Notice that when the boron is less than 2.5% the brazing-temp line is no longer in the “liquid” section of the chart, but is now situated in the “Slush” zone between the liquidus and solidus lines! When this happens, the BFM will begin to solidify, even though the brazing temperature is being held constant. Isothermal solidification (ITS) has begun!

Please note that ITS will not occur by merely holding the BFM at brazing temp for a few minutes. Instead, it requires much longer times at temp, typically 30-minutes minimum, and sometimes as long as one or two hours, or even longer, depending on the size of the furnace load. Experience will indicate the time required, depending on the mass of the parts, how much BFM is present, and thus, what clearance is being used in the joint.

Note: Wide gaps are not effective for trying to implement ITS. For good ITS results, the gaps should be tight at brazing temp, typically 0.000-0.003” (0.000-0.075 mm), and the quantity of BFM applied should be just enough to fill the volume between the faying surfaces of the joint. Not only can thick braze-joints be a deterrent to effective ITS, but excessively large amounts of applied BFM can result in large, heavy braze-fillets which, because of their large size and volume, can prevent ITS from occurring in those external fillets, even though their re-melt temperatures may have actually increased a little because of some boron diffusion.

In summary, if the gaps are large, or if too much BFM is applied, there will typically be too much BFM present in the joint area to be able to effectively diffuse away enough of the boron temperature-depressant, and ITS will not occur, even for lengthy holds at brazing temp.

Please note, too, that because of the much larger size of silicon and phosphorus atoms used as the temperature-depressant in many other nickel-based BFMs, ITS will be very difficult, if not impossible, with those BFMs using silicon or phos as the temperature-depressant additives instead of boron.

Problem solve and improve with our technical experts.