Proof of Carbon Migration

[Steve Sells]

I have brought up this topic a few times in the knife chat, Here is a simple to under stand paper, with photos, short and to the point. http://asmcommunity.asminternational.org/static/Static%20Files/IP/Magazine/AMP/V167/I02/amp16702p24.pdf?authtoken=cb31535fc425fd4bcb22e182921e2963ed861895 This not only show carbon migration in the first fold, but also to clear up the the fact of the addition of even 4% nickel to an alloy is not enough to stop migration, While nearly pure nickel will, the roughly 3.5% content of A203E is less than is needed for that.

Edited March 10, 2009 by steve sells

[Quenchcrack]

OK, I need Jim Hrisoulas to explain this to me. The use of acid to etch a metallurgical specimen dissolves the IRON and leaves the CARBON replica in place. High carbon areas etch dark and low carbon areas etch light. If the carbon diffuses and homogenizes, what causes the banding? Microstructure? Pearlite is iron carbide in a ferrite matrix and it etches dark. Ferrite has basically zero solubility for carbon so it etches light. But in a pattern welded blade, the carbon content is the same all over so there should be no inhomogeniety in the iron based structure. It should all etch the same. And if the carbon diffuses into the Nickel, why does the nickel still etch white? Inquiring minds want to know.

[dablacksmith]

ok that was a interesting read! it explains why a nickle damascus shows much better than a non nickle ... but why if the carbon is the same (and therefore hardness ) does the non nickle damascus even show a pattern in a etch? what is the difference ? why does the acid react differently to one section than the other? i know a mild /wrought iron and hi carbon blade will show pattern in a etch ...why?

[Steve Sells]

OK, I need Jim Hrisoulas to explain this to me. The use of acid to etch a metallurgical specimen dissolves the IRON and leaves the CARBON replica in place. High carbon areas etch dark and low carbon areas etch light. If the carbon diffuses and homogenizes, what causes the banding? Microstructure? Pearlite is iron carbide in a ferrite matrix and it etches dark. Ferrite has basically zero solubility for carbon so it etches light. But in a pattern welded blade, the carbon content is the same all over so there should be no inhomogeniety in the iron based structure. It should all etch the same. And if the carbon diffuses into the Nickel, why does the nickel still etch white? Inquiring minds want to know.

Carbon does not diffuse into or bond with nickel as it is not a carbide former, but 204E is not nickel , its a steel alloy that contains some 3.5% nickel tho, there is a difference.

[Steve Sells]

ok that was a interesting read! it explains why a nickle damascus shows much better than a non nickle ... but why if the carbon is the same (and therefore hardness ) does the non nickle damascus even show a pattern in a etch? what is the difference ? why does the acid react differently to one section than the other? i know a mild /wrought iron and hi carbon blade will show pattern in a etch ...why?

Because carbon isn't the only thing that effects etching. High Mn content when etched is black, so in an 1010/1095 mix its the Mn making the difference in the layer colors. also WI has silicon, there are many things that react, or protect from etchant 's action. carbon is only one of those. even a mono steel ., like Tamahogane or cable, has the exposed areas where carbon is lost to atmosphere. more complex than a simple 2 page report can explain, but this is short enough for most people to follow.

[Quenchcrack]

Carbon does not diffuse into or bond with nickel as it is not a carbide former, but 204E is not nickel , its a steel alloy that contains some 3.5% nickel tho, there is a difference.

OK, the 204E is not nickel, it is a low carbon steel. That means the carbon will diffuse into the steel part as you state. After forging it has the same carbon as the W2 (which is now lower since it is donating steel to the lower carbon areas). So if the nickel is substitutionally alloyed into the 204E it will be randomly and uniformly distributed throughout the iron matrix, not in bands. If nickel and carbon are randomly dispersed through the 204E and carbon is the same as the W2, what causes the light/dark bands?

[ThomasPowers]

The nickel makes the alloy resist etching more. Ni does not migrate very much and so you should get alternating bands of Ni alloy steel and non-Ni alloy steel that causes light and dark bands.

[Steve Sells]

I said nothing about Nickel migration, nor any other of the more common alloys. So I will now state that while they do migrate/diffuse, they do so at such a slow rate, they basically stay in the layers they started in. While the carbon moves quite fast, as the steel gets hotter.

The contents of steels vary with each mill, and each batch vary a bit also. So many report a range of contents, which is why its recommended to get a mill sheet for each batch we get.

But generally this test in question used a sample of A203E with C 0.14 Mn 0.58 Ni 3.43. W-2 has no nickel lending that layer a dark color, while the Nickel of the A203E shows no appreciable diffusion, therefore remaining in the A203E layer leaving it bright.

Edited March 11, 2009 by steve sells

[Quenchcrack]

Actually, Volume 9 of the Metals Handbook, published by ASM, states that nickel alloys up to 96% nickel can be readily etched with a mild mixture of HCl and Ferric Chloride. I would assume A203 would respond quicker than, say Nickel 200 or 270. I thought Ferric Chloride was the etch preferred for etching pattern welded blades? As for theW2 having no nickel, thats true. But the A203E and the W2 must have the same carbon content due to diffusion and the carbon is what gives color to the etch. So if the carbon content is now uniform, and the nickel does not pose a serious resistance to etching, where do the light and dark bands come from?

[Steve Sells]

where did you get the idea that nickel does not resist etching , or that only carbon can etch?

if that was true, then forget most stainless steel pattern welding. I assume all steel can be etched with enough time to a strong enough acid. That is not the point, as I thought you understood, but I have been wrong before.

Edited March 12, 2014 by steve sells

[Tom.K]

Interesting paper, thanks for sharing.

[raker]

Carbon will migrate in steel when heated to above the critical point and the carbon goes into solution. The depth of migration will depend on the time at temp. In damascus or pattern welded steel, the layers get pretty thin and you have migration from both sides. This will enable the carbon to be completely diffused and uniform through out the billet. That said, on a san mia billet that has not been above critical very long, the carbon will only be able to migrate a few thousandths and the core will not loose much of the carbon, percentage wise.
On etching the billet, all steels will be differently affected by the acid. This is what causes the different layers to show up as bands. The composition of the steels will determine what the steel looks like after it has been etched. Even the boundries will show if they have been exposed to the air. Cable and a stacked billet of a singel steel will show boundry lines. The boundries etch different because they have a different composition than the subsurface of the steel.
Something you might try is to use 4 to 5 different steels for a damascus billet and after sanding smooth, slowly heat the steel. You will notice that they will all produce different oxidizing colors. This is due to the different elements in the steel.
The importance of understanding carbon migration is to make sure you know the carbon content of the end product. If it is to be used for hardware, the amount may not be too important, if it is to be used for a blade, it may be very important

[ThomasPowers]

What we are dealing with is DIFFERENTIAL etching---why using weak solutions are often better than using strong solutions of etchant when etching Damascus. (a common set up is to cut the PCB etchant you buy at RadioShack, Ferric Chloride, to 1/3 to 1/4 strength.) It all etches but at *different* rates and yes the higher Ni content does seem to be the layers that etch the least as given by the topology that forms.

Carbon is NOT the only thing providing colour---as mentioned previously Mn also has an affect as does other alloying elements.

So the light and dark bands come from what we have already mentioned several times above: layers of steel that have different alloying elements at different concentrations that differentially etch. It is not just an effect of carbon content or Ni content for that matter but the whole slew of stuff that is in the steels used.

[dablacksmith]

ok so the gist of the matter is that for hardening perposes the carbon migration is inportant but for the etching purposes less so... and a billet of hi and low carbon will end up a med carbon...that does help explain some failures in the past... and why i always harden knives and swords in oil...thank you for that article!

[Quenchcrack]

That was a heck of a dialog considering it all started over the definition of Damascus steel from a now-defunct Japanese Blade site!

[ThomasPowers]

It has been surmised that some of the toughness folks report from pattern welded blades may be due to them having a lower carbon content than was expected due to the dilution of the higher carbon with lower carbon.

Of course you can always use 2 or more different high carbon steels and no low carbon steels---some makers call that a "super damascus"...

[Torin]

Of course you can always use 2 or more different high carbon steels and no low carbon steels---some makers call that a "super damascus"...

Well, I don't know about "super", but I like 15n20 and 1095 for that reason. Both high-carbon steels, so the heat treating is a bit easier.

[minaraimono]

Thanks for the article. I do wish it had about ten times as much text, but its more or less cured me of any doubt that the carbon content is homogenous in pattern-welded steel.
I am, however, still really curious what is responsible for the visual layering.

Just saying, "The compositions are different" hardly stands up to the scientific integrity set forth by this article. I think we owe it to ourselves to get to the bottom of it. And if there isnt anyone here who can explain this yet, maybe it will be the subject of next years Jacquet-Lucas award?

I make billets of damascus out of mild/wrought iron (its carbon content is lower than standard US "mild steel" but I've never seen a spec sheet on it) and Japanese Shirogami (white) #2. The white steel is known for its high purity. It has 1% carbon, and practically nothing else. Mn, for example is .20~.30 wt%. Si is .10~.20 wt%.
And seeing as Sulfur is less that .004%. and they dont even list Nickle, I think its safe to say there is'nt much.

After forging and HT I don't even etch it in acid. I sand blast it. And that reveals the laters. I would just like to understand that a bit better.

[johnptc]

nice article with photos

Edited September 27, 2013 by steve sells
clean up

[Steve Sells]

As many have noticed, the link form the initial post is dead.  ASM reformatted their web site, and as a result the link is invalid.

 

I got permission from Sue Sellers, at ASM to post the PDF at IFI for our use so all can review the information in the discussion above.

 

Jacquet-Lucas Award: Metallography of a Modern Pattern-Welded Steel Knife Blade
Thomas Nizolek
Advanced Materials & Processes, Volume 167, Issue 2, February 2009 (ASM International)
www.asminternational.org

amp16702p24.pdf

[Rich Hale]

Tomorrow morning before breakfast I will take a small pile of knives and assist them  in migrating first to California..then to wherever they wish to reside. i do not plan on doing a follow up paper on this but I suspect this migration can take place with minimal heat.

[Glenn]

This is the beauty of IForgeIron.

 

Buried in the archive back in 10 March 2009  to 15 March 2009   we have information that has been available to all and now some 5 years to the day it was found again. Thank you Steve for taking the time to get permission to post the dead link information back to IFrogeIron for all to use as a reference.

[jmccustomknives]

Here's a neat little visual of carbon migration.  I forged this knife from band saw blades and pallet banding.  Since the bandsaw was fairly low carbon (.70 +or-) and the pallet banding didn't harden I thought it would be good to slip a Black Diamond file in the center on the last weld.  For some reason the layers overlapped and enclosed around the file making it the core.  You can see along the edge at the bottom how it is dark and fades out into the rest of the blade.  The layer cross through this.  I'm pretty sure the carbon migrated that far in the little time it took me to weld and draw the blade out.  The differential etching took place on both sides, almost mirror. 

[mitch4ging]

Steve & JMC,
Thanks for the great information and visual!

Rich, thanks for the grin, I have a few I would like to migrate out of state! : )

[Ric Furrer]

I like studies, but this dog has hunted a few times before.

 

John Verhoeven & Howard Clark, (Carbon Diffusion between the Layers in Modern Pattern-Welded Blades, "Materials Characterization". Volume 41, Issue 5, November 1998, pg 183-191)

 

And

Byron Skillings article for the ASM

http://www.bladesmithsforum.com/index.php?showtopic=15749

 

I think this issue has been studied enough...it is a matter of knifemakers diffusing the info.

 

Ric