Measurement of the various potentials in nitriding or nitrocarburizing atmospheres and the principles used by these systems have been well known for years. However, there are obviously some big differences in the way such instrumentation is designed and how they behave in regular industrial furnaces compared to laboratory environment.
|Fig. 1. Potentials by process|
We all know that the foremost task in a nitriding application, besides a perfect control of the temperature, is the control of the nitriding effect of the atmosphere. This can be done in various ways and is mainly dependent on the specification to which you are required to treat your parts.
No specification: Proper nitriding conditions are set using fixed gas flows.
SAE AMS 2759/6A: Atmosphere process is controlled in % dissociation in one or two stages depending on steel, case and required process.
SAE AMS 2759/10: Atmosphere process is controlled in units – nitriding potentials – instead of the dissociation.
There are typically no specifications for controlling the nitrocarburizing process. However, the more accurate and modern systems control the nitriding and carburizing potential, just like in a nitriding process. The carbon-bearing gas that is added is CO2 in most cases. But times are changing.
SAE AMS 2759/12: This fairly new specification demands concurrent control of two potentials – nitriding potential (KN) and the carburizing potential (KC) – that must be specified as either KCB, according to the Boudouard reaction, or KCW, according to the heterogeneous water-shift reaction.
There are other processes, like oxi-nitriding, not yet defined in an official specification but already in use. This process mainly targets treatments of higher-chromium parts. The oxygen is used to break the passivation layer we typically find on such steels. In an oxi-nitriding process we have to control KN and the oxidation potential KO at the same time.
And there is something more to consider. There are two new control variants that are aiming for specific properties of the compound layer, which might be better known as the white layer.
|Fig. 2. Thermal conductivity analyzer and a modified oxygen probe|
Potentials and the Processes
First, let’s have a look at the potentials we have to control to fulfill the various specifications (Fig. 1).
KN is quite easy to determine. The only things we need to know are the partial pressures of ammonia and hydrogen and the nitrogen activity (aN). In order to determine the dissociation, it’s sufficient to know the ammonia partial pressure.
This completely changes the moment we have to deal with a nitrocarburizing atmosphere. In order to control the carburizing effect, which might be better described by the carbon activity, aC, we must deal with various KCs depending on the carburizing gas we add to the furnace atmosphere and their reactions.
KO and the oxygen activity (aO) can be derived from the partial-pressure ratio of water vapor and hydrogen.
There are many measuring systems that can be used to measure the partial pressures or changes in the atmosphere composition, like pressure transducers, infrared or ultraviolet analyzers, or hydrogen-permeable membranes, but we will focus on the use of a combination of a thermal-conductivity hydrogen analyzer and a modified oxygen probe (Fig. 2).
Now we will briefly present how the signals of these two measuring devices can be used to determine the potentials we have to control – KN, KCB or KCW, and KO.
In the following equations to the left, “s” is the dissociation degree.
... with KW being a thermodynamic constant depending on temperature to determine the total gas composition in the exhaust.
According to SAE AMS 2759/6A, dissociation equates to 100% minus residual ammonia. This is exactly the value you would read on a burette if you read it upside down. Note that this value is not the same as the dissociation degree (s) shown in the equations above but it can easily be converted by: D=2s/(1+s)
In order to determine “s,” we use the hydrogen analyzer, which is giving us an approximation for H2[%]. In combination with the millivolt signal of the oxygen probe being a measure for the ratio of pH2O/pH2, specifically H2O[%]/H2[%], we get the amount of water vapor H2O [%]. The total hydrogen that originally came out of the dissociation of the ammonia is: Total H2 = H2[%] + H20[%]
With the total H2[%] being the same as 1.5s/(1+s)*100%, we can derive “s.” As the oxygen probe signal is also giving the partial pressure ratio of pCO/pCO2 and as we know KW, which is a function of the temperature, we can derive both KCW and KCB if the gas flows are known.
All of the above is well known and not specifically new. So, let’s finally have a look at something new.
|Fig. 3. Lehrer diagram with Kc controlled to zero (left); three-dimensional graph at fixed temperature at 575°C (right)|
Phase Control and Weight Percentage Control
As stated, both control variants aim for specific properties of the white-layer structure.
The weight-percentage control uses concurrent KN and KC control in order to create a given weight percentage of nitrogen and carbon in the compound layer. There are a few control policies depending on the gas mixture provided, but we recommend the use of ammonia versus dissociated ammonia to maintain the proper nitriding potential and CO versus CO2 to maintain the proper carburizing potential. Both potentials will be controlled in such a way that the working point stays on the nitrogen and carbon iso-concentration lines in the applicable Fe-N-C phase diagram (Fig. 3).
|Fig. 4. Micrographs show how white-layer porosity is influenced by replacing nitrogen with carbon for 1010 steel|
Please note that in the left Lehrer Diagram, KC would be controlled to zero, and in the right diagram, we control the two potentials on a working point that is represented as a three-dimensional line through a set of diagrams, each varying over temperature.
Figure 4 shows how the porosity of the formed white layer can be influenced by gradually exchanging the nitrogen content with carbon.
|Fig. 5. Ferric nitrocarburizing parameters (KN 8.497, KC 0.032) result in high nitrogen and low carbon – different alloy results shown (530°C).|
Figure 5 shows how this affects different steel grades. The first photo shows high KN and low KC, or high nitrogen and low carbon, whereas the second one results in higher carbon and less nitrogen. While the compound-layer thickness stays fairly constant, the porosity gets lower with higher carbon amounts.
Another reason for a weight-percentage control is given by the carbon distribution within the white layer. If we look at a regular nitrocarburizing result, the typical profiles of nitrogen and carbon with a hump in the carbon profile mark the boundary toward the diffusion zone.
|Fig. 7. Carbon, nitrogen and oxygen profiles|
This effect is also known as an uphill diffusion of carbon during nitriding (Fig. 7). Figure 8 shows how weight-percentage control can avoid this effect and result in much more uniform layers. Hence, the properties within the layer do not change with increasing depth.
The second control variant is called phase control. This variant does not control the working point to stay on the iso-concentration lines, but it controls KN and KC in such a way to keep aN and aC at a level to match the activities of the (carbo-)nitride that is aimed for throughout the process (Fig. 9).
|Fig. 8. Weight-percentage control avoids up-hill diffusion.|
This can be best explained as controlling the nitriding potential parallel to the phase boundaries. This control variant enables the treatment of extremely dense and compact white layers.
In Figure 10, the top micrograph shows a conventionally nitrocarburized part, and the bottom micrograph shows a phase-controlled part. Both parts have been treated in the same industrial furnace in a regular workload. We can immediately see that the phase-controlled variant does not show the apparent porosity of the conventional process.
|Fig. 9. Control of the nitriding potential parallel to the phase boundaries|
Obviously, the control equipment you need in order to perform such processes has to have certain properties, including:
- A gas panel or cabinet equipped with NH3 versus dissociated NH3 for the KN control and CO versus CO2 for the KC control (Fig. 11a).
- A sensor system comprised of a hydrogen analyzer (Fig. 11b) and an oxygen probe (Fig. 11c).
- A programmable controller (Fig. 11d) able to determine the required potentials and able to control several potentials concurrently while providing an compatible recipe.
|Fig. 10. Conventionally nitrocarburized part (top); phase controlled part (bottom)|
For more information, contact Pat Torok at United Process Controls, 8904 Beckett Road, West Chester Ohio 45069; tel: 1-800-772-1000; e-mail: Pat.Torok@group-upc.com; web: www.group-upc.com
Additional related information may be found by searching for these (and other) key words/terms via BNP Media SEARCH at www.industrialheating.com: nitriding potential, oxidizing potential, thermal conductivity hydrogen analyzer, oxygen probe, phase control, dissociation
|Fig. 11. Control equipment|