Carbon diffusion before and after the weld?

BrianThie

BrianThie

Joined
Jun 18, 2007
Messages
382
Hello all

I have a question concerning carbon diffusion that has been bugging me for a while and I cannot seem to find an answer in the reference material I have checked.

Let’s say for this argument that we are starting with 2 pieces of steel – one high carbon and 1 low carbon. If these pieces are dry forge welded together, at what point does carbon diffusion begin. It would seem logical to me that diffusion would occur as would be normally expected, but be suppressed somewhat by the fact that the 2 pieces are indeed separate, but still touching. After the weld is set, it would seem to me that carbon diffusion would occur unrestricted.

So my question is: Does the rate of diffusion differ from 2 pieces of steel just touching one another as compared to 2 pieces actually welded together and if so how much?

Thanks for your input
Brian
 
Brian, I am no expert, but what I noticed with my san mai was that at the first high heat weld some carbon diffusion occured, but the low temperature of forging thereafter not much diffusion occured. I think time and temperature go hand in hand, long time at low temperature and short time at high temperature will cause the same diffusion.
 
Yes Sam, you are correct. Carbon diffusion is a function of time, temperature and distance. The higher the temp, the faster the diffusion.

What I am looking for is how the diffusion rate is affected before the weld is made--while the pieces of steel are still seperate.

To go to extremes, if the pieces were at 2100 degrees each and 1/8" apart, there would be no diffusion, but if the pieces were moved together until they touched, would diffusion take place at the same rate as if they were welded or would the ever so small gap between the 2 pieces supress the diffusion.

I am having trouble putting words to this so I hope it is making some sense. I am also basing this on the assumption that we are in the absence of scale ect.

I guess another way to ask this is the rate of diffusion between 2 pieces of steel that are just touching the same as 2 pieces of steel that are welded together - assuming the temperature remains the same?

Brian
 
I think that because diffusion happens on a molecular level, there needs to be a molecular bond between the two materials for there to be a bridge for the carbon to travel across. If they are touching but have not welded at that temperature then I don't think it can happen.
 
Last edited:
Sam - I think that diffusion can occur with two pieces just touching. I think that this is similar to iron taking carbon out of the atmosphere in a bloomery. Diffusion to homogenous steel is going to take longer, of course, since there is a greater distance to travel. I think that surrounding the low carbon steel with high carbon steel will decrease that diffusion time. I don't know if the rate of diffusion would be slowed by them simply touching or if it would be comparable to welding the two pieces.
 
Last edited:
Diffusion in the bloomery process happens due to the very carbonizing atmosphere and the sheer abundance of carbon going every where, I think one carbon rich peice next to a carbon poor peice in a neutral atmosphere the carbon uptake to the other would be VERY slow. It is intriguing, I wish I had a temperature controlled oven to do a test! Keeping them from welding would be tricky.
 
Diffusion occurs because of a concentration gradient between two different materials, and usually requires some sort of actual contact. While it is a function of both time and temperature, temperature is a much bigger factor than time. And as mentioned, distance plays a big role as well. If their is some distance between two materials, the energy necessary to cause atoms of carbon to jump from one material to the other would be huge, and not nearly as likely to occur.

So as to your original question Brian, I'm pretty sure that the diffusion rate of the carbon will be much slower in pieces that are just touching than in steel that has been welded. My reasoning is this: even if you machined all your pieces of steel, surface ground them to a very fine finish, you will still not have 100% contact with the surface of the other steel. There will still be surface textures and gapping that will occur, and while these may be very very small, in comparison to the atomic level, they are significant enough to slow diffusion and there will still be spaces between them where contact does not occur and carbon will not want to diffuse from the higher carbon steel into the lower steel. Whereas if you have forge welded the steels together, there are bonds created between both steels and (hopefully, assuming no bad welds ;) ) you have 100% contact (so to speak). Also with the way diffusion usually occurs (little point defects creating vaciencies, that allow atoms to change lattice points, and the vacancies created by the shifting allow atoms to move into the vacancies, be shifted around and move inward), this should be much easier to do when there is an atomic bond between the mateirals, rather than just surface contact.

In terms of actual rate differences between the two methods, that would probaby be quite difficult to calculate, and would probably varry greatly depending on what type of finish you put on the steel and how precise you were with the fitting/gapping.

And a note on carburization based on carbon heavy atmospheres/direct contact with carbon (like case hardening compounds, etc). Generally with this you are dealing with free carbons, so the diffusion rate will be a bit faster than when working with carbon that is already bonded and integrated within the crystal lattice of steel. So this carbon will be much more likely to diffuse into a lower carbon steel based on contact alone (plus the nature of carbon rich atmospheres, or other sources of "pure" carbon is such that surface contact with the steel is a bit easier to achieve, being in fine particles).
 
Last edited:
Graham

Welcome to the forums!

I agree with you that the difussion rate would be slower with pieces that are just touching, but this speculation based on my reasoning.

I guess what I am looking for is does anyone know of a source that has documented this? An actual proof of a theory.

My best sources for stuff like this is John D. Verhoeven's publications and he doesn't address this question from what I can find.

Brian
 
Diffusion is also dependent on the surface area exposed, which in this case is related to the amount of the different steels actually in contact with each other. Also, remember from Verhoeven that diffusion of carbon depends on a gas. Iron surrounded by graphite in a vacuum wont diffuse into the iron. It has to go to the CO & CO2 mixture, then into the steel, or not, depending on the concentration of the gasses. This applies to the 2 steels separated by some distance, however small. There has to be oxygen for the carbon to combine with and leave one steel to be carried to the other. So you have to have a decarburizing atmosphere for the higher carbon steel and carburizing atmosphere for the lower carbon steel, at the same time:eek:.
 
Nature always wants to reach equilibrium.The higher carbon side will drive that carbon into the lower carbon side.For steel nothing will happen until you get above critical and of course the higher the temperature the faster the diffusion. The grain boundaries have higher energy states so there is more activity and thus more diffusion.The strangest carburizing I've seen is in carburized powder metal parts.The part had a carbon gradient but so did each particle !!
But to the original question .The diffusion will occur when two pieces are just touching but there are a number of variables so it would be hard to predict.With welded steels you could get some numbers. Of course diffusion rates would also depend on the alloy and nickel containing alloys would have lower diffusion rates. Carbon with it's small size will diffuse much faster than other elements.
 
You must log in or register to reply here.
Back
Top