Differential Density

[Justin Gifford]

I followed a link to Angel Sword's site that somebody posted in another thread. In their definition of a "true sword" they make brief mention of differential density.

Differential Density – Some smiths believe in this technique, while others just as strongly claim that is has no basis or value. In any case, the concept is that the edge of a blade can be packed more densely than the body through hammer blows, thus giving it additional strength and hardness.

I don't mean to criticize or praise AngelSwords; I've never used any, or even seen any in real life, for that matter. I'm just curious about this differential density. I've never heard of it, so I wonder what any of the smiths around here have to say about it.

~JAG

[James Roberts]

As a scientist with a little metallurgy under the belt, I would say that differential density could well be possible- a lattice of metal with little carbon content will be denser than a highly carbonated (?) lattice. But when we discuss swords, the differences between one section of the blade and another are more likely to be the result of the changes in the actual structure of the steel. There are a lot of proven swordmaking techniques that have not so far been understood, but this is one claim that is very difficult to test. Rather just say that the edge has been hardened through hammering, or "differentially hardened".

[Michael Stora]

Originally posted by Justin Gifford
I followed a link to Angel Sword's site that somebody posted in another thread. In their definition of a "true sword" they make brief mention of differential density.



I don't mean to criticize or praise AngelSwords; I've never used any, or even seen any in real life, for that matter. I'm just curious about this differential density. I've never heard of it, so I wonder what any of the smiths around here have to say about it.

~JAG

Even Angel Swords says something like "some say" and not "we say" which is a darn good thing too . . .

I have a combined doctorate in materials science and engineering and say it is hogwash.

Any time you strain an elastic material, there is a small volume change because Poisson's ratio is not perfectly matched to constant volume, but this would have no effect with respect to the performance of a sword beyond that of the built-in strain. Now an edge under compressive strain is tougher than the edge of the same hardness without compression. That might be what they are alluding to without knowing it.

Mike

[C Taylor]

What they might be reffering to is the differential tempering done by japanese smiths, in which the blade of the sword has a higher carbon content than the belly of it, makeing the edge denser and harder, and holds an edge better. While the belly is flexible and soft.

[Michael Stora]

Originally posted by C Taylor
What they might be reffering to is the differential tempering done by japanese smiths, in which the blade of the sword has a higher carbon content than the belly of it, makeing the edge denser and harder, and holds an edge better. While the belly is flexible and soft.

The Martensite/Cementite edge has a lower density that the pearlite core. Body centered tetragonal is more loosely backed that body centered cubic (not to mention that carbon is a much lighter element than iron).

It is, however, under compressive stress which makes the edge tougher than would normally be the case with a straight sword.

Mike

[Eric Litton]

I always thought that Japanese swords were clay coated so that the heat treatment would cool faster on the edge than the back, giving it a hard edge, but that the edge and back had the same amount of carbon in them.

Originally posted by C Taylor
What they might be reffering to is the differential tempering done by japanese smiths, in which the blade of the sword has a higher carbon content than the belly of it, makeing the edge denser and harder, and holds an edge better. While the belly is flexible and soft.

[C Taylor]

yes, it mainly has to do with heat treating, but when the steel is heated, the crystalin srtucture expands and more carbon enters. The edge of the blade cools quicker due to the lesser amount of clay, and more carbon is trapped in it. The belly cools slower and carbon atoms are more likley to escape. Japanese swords are also folded so there are alternating layers of soft and hard steel (A.K.A low content and high content carbon steel respecfully).

[W. Bogdon]

Originally posted by C Taylor
Japanese swords are also folded so there are alternating layers of soft and hard steel (A.K.A low content and high content carbon steel respecfully).

This isn't quite right - the Japanese typically chose multiple chunks of steel, each of very similar carbon content, for any given blade. The folding of these chunks was done to evenly distribute impurities throughout the blade - thus reducing the chances of them remaining clumped together at a single weak point.

-Craig

[C Taylor]

well, that's what i meant.

[Kevin R. Cashen]

This is one of those bits of nonsense that just will not die. I don't know how many times we have laid it in it's grave in a bloody heap on this forum alone, just to have it sprout another hydra like head and start breathing bull$&!# all over again! So none of my intensity is directed at anybody on this forum, but at all of those folks on the internet and beyond that don't at least crack a book or two before perpetuating this utter nonsense.

Steel is not compactible on this level, that is why forging and rolling mills work. You squeeze it one way and it will expand the other, not just get smaller. You don't even need a metallurgical background to figure this one out, high school level science will do. You can't pack metal molecules without the proper preparation. Preparation like the boys at the trinity test site had.

Now for the basic metallurgy. If you hammer on it at a proper forging temperature, recrystalization will fix your meddling as you go. If you cold work it or plastically deform it below critical (ala "edge packing", I hate to even say it ) due to vacancies, edge dislocations and other havoc in the lattice the steel will probably be less dense, yes the steel would probably have a slightly lower density.

Fortunately most all of it can be corrected on the very next Ac1+ heat. In most cases, heat has the ability to correct or destroy almost anything the hammer does, and thank god for that!

[Justin Gifford]

Originally posted by Kevin R. Cashen


You squeeze it one way and it will expand the other, not just get smaller. You don't even need a metallurgical background to figure this one out, high school level science will do.

Kevin, thanks for the concise rant! I suppose that explains why I haven't heard of it... The excerpt above is what I'd initially thought when I read it at the AS website, but I figured I'd check.

Thanks again! *Ducks in mortal fear of Hephaestus' fury*

~JAG

[Al Massey]

...does have a slightly higher specific gravity than iron, I believe, in other words the same volume of steel will be very slightly higher in mass than that volume of iron. Some Japanese warriors believed a sword with an iron core would be lighter, but I'm not sure if the difference in weight would be readilly noticeable,unless there was a huge welding flaw between the core and the rest of the blade (which there often was!).
Basically, regarding forged steels, with modern smelted, rolled alloys you're not going to see a "higher density" between forged and non-forged blades. Perhaps this was the case with steels produced hundreds of years ago, which were far less homogenous.

[T. Karlsson]

Originally posted by Kevin R. Cashen
Fortunately most all of it can be corrected on the very next Ac1+ heat. In most cases, heat has the ability to correct or destroy almost anything the hammer does, and thank god for that!

Does that mean that you won't have any of advanteges from cold forging? Like increase in dislocation and orientation of the material in the direction of stresses?? Increase in dislocation makes steel harder, but maybe it makes the steel very brittle??

To forge/hammer the edges more than the bulk should make them harder than the bulk due to the contribution of new dislocations wich make dislocation movement more difficult.

[Al Massey]

...because of the style of production, old steel often was very porous, with "air spaces" and forging improved the quality. This explains the advice given in one "technical manual" on weapons production that iron and steel items had to be forged instead of being cut and ground/filed if strength was expected.

[A. Ko]

Originally posted by James Roberts
As a scientist with a little metallurgy under the belt, I would say that differential density could well be possible- a lattice of metal with little carbon content will be denser than a highly carbonated (?) lattice. But when we discuss swords, the differences between one section of the blade and another are more likely to be the result of the changes in the actual structure of the steel. There are a lot of proven swordmaking techniques that have not so far been understood, but this is one claim that is very difficult to test. Rather just say that the edge has been hardened through hammering, or "differentially hardened".

I mentioned this in response to another thread. They basically hammer the edge to work harden the steel at a colder temperature, believing in what Kevin Cashen (ABS Master Bladesmith) points out as a misconception known as "edge packing". Some bladesmiths believe that by hammering on the edge you're increasing the material density and they argue that when they break the blade to do a sectional analysis the crystaline growth is finer than the body.

Howard Clark (also ABS Master Bladesmith) and Kevin Cashen have countered that if you normalize the blade and do a proper heat treating to the blade, you can get that same fine grain growth throughout the entire blade, and such would cause the blade to function better.

My personal feeling is that Howard and Kevin know a great deal about heat treating and know what constitutes a performance sword. A sword made under misconception or a lack of understanding of metallurgical concepts does not produce a structurally optimal sword even though it may feel "alive" or may cut rope, etc.

Having proper crystaline structure is far more important than trying to hammer the edge at colder temperatures to achieve material density. One is arrived by a heat treating process, the other is by physical labor. The former produces a superior sword. The latter basically destroys any concept of the sword marketing as functioning as a "single crystal" because the lattices no longer line up - as if they ever did in the first place.

What I see here is marketing myth and misconception over producing proper metallurgical knowledge, and this could potentially (note: "potentially") produce a suboptimal sword.

[Kevin R. Cashen]

Originally posted by T. Karlsson


Does that mean that you won't have any of advanteges from cold forging? Like increase in dislocation and orientation of the material in the direction of stresses?? Increase in dislocation makes steel harder, but maybe it makes the steel very brittle??

To forge/hammer the edges more than the bulk should make them harder than the bulk due to the contribution of new dislocations wich make dislocation movement more difficult.

all that you are saying is applicable to cold working alone. Take any metal with a recrystalization temperature that is above room temperature (lead, and certainly mercury, won't work so well) and cold working will induce stress via slip, twinning, dislocations etc.. to the point that continued plastic deformation could result in failure. This is why we frequently anneal nonferrous while cold forming it with hammering. But with that anneal, recrystalization occurs and the whole slate is wiped clean.

Steel can be work hardened like other metals but upon heating to Ac1 you will be right back to where you started. Well, ok, perhaps not exactly back to where you started, since the strain induced by the coldworking will increase the points of nucleation for the newly forming austenite grains.

But you see this is the ultimate irony of the whole "hammering to refine grain" thing, in the end it is not hammering that makes the smaller grains it is the heat to Ac1. The hammer can increase the strain that drives the process but nowhere near as efficiently (evenly) as what some microstructres can, if done right. So once again the fire wins out over the hammer.

In fact I think a lot of these notions came about from mistakenly applying nonferrous metallurgy to the working of steel. In bronze or copper, for instance, cold working can be the only driving force behind recrystalization, since you don't get goodies like martensite, bainite or pearlite.

[Al Massey]

...there is some evidence that many swords (ferrous) were "work-hardened" by cold hammering as opposed to heat-treating to harden the blade. Depending on the carbon content, as well as other trace elements in the metal, cold hammering may be more effective as a means of hardening. A sword examined by metallurgists in Oakeshott's "Records" apparently showed considerable evidence of cold-working. But this would likely be done as an alternative to heat-treating, not in addition to it, imho.l

[T. Karlsson]

What you are saying, Kevin R. Cashen, makes sense, but i found something interesting in a book. The books name: Deformation and fracture mechanics of engineering materials, writen by Richard W. Hertzberg.

"Engineers have found that the fracture resistance of a forged component can be enhanced considerably when the forging flow lines are oriented parallel to major stress trajectories and normal to the path of a potential crack. As such, forged parts are considered to be superior to comparable castings because of the benefits derived from the deformation-induced microstructural anisotropy. Japanese armor makers took advantage of this fact in their manufacture of the samurai sword. They heated a billet of iron to an elevated temperatur where it was folded back upon itself by repeted blows of forging hammers and then placed back in the furnace untill ready for another forging and folding operation. This was done 10 to 20 times, resulting in a sword blade containing 2 to the power of 20 layers of the original billet. After masterful decoration and a special heat treatment , this aesthetically appealing and structurally sound weapon was ready for its deadly purpose".

Here it seem that the forging is a real benefit even if your going to do "special heat treatment"(what ever that is) upon the steel afterwards. Maybe its just that you will get the inpuritys alinged with the direction of stress and there for they will not act as crack startpoints as much as they would of otherwise??

[Kevin R. Cashen]

[QUOTE]Originally posted by T. Karlsson
[B]

But there is a big difference between the lines of slippage within the grains and the flow lines left from the original casting and rolling operations. The very microscopic lines of slip will be completely erased by subsequent heating.

The "flow" lines in steel which are quite often visible with the naked eye are stringers of inconsistancies left from the casting process, which have been forge (rolled mostly) in order to break them up and align them lenghtwise in the stock. And yes it is pretty much the same technique used by the Japanese to turn a slaggy mess into steel. If you cut a strip from a sheet of steel that is 90degress to the flow lines, yes it will be weaker than one that was rolled to length. It is the classic forged crankshaft example that one sees in all of the basic metallurgy books. But it has virtually nothing to do with our conversation on "edge density". All the steel we buy today has already had the rolling/ forging/aligning process done for us.

[Kevin R. Cashen]

Originally posted by Al Massey
...there is some evidence that many swords (ferrous) were "work-hardened" by cold hammering as opposed to heat-treating to harden the blade. Depending on the carbon content, as well as other trace elements in the metal, cold hammering may be more effective as a means of hardening. A sword examined by metallurgists in Oakeshott's "Records" apparently showed considerable evidence of cold-working. But this would likely be done as an alternative to heat-treating, not in addition to it, imho.l

No argument here, but no wrench either. If one is dealing with a mostly ferritic material without enough carbon to reach any appreciable hardness, cold working will provide much more rigidity than any heat treatment would. This is would be no different than with bronze. But if somebody wants to challenge a properly heat treated eutectoid steel, with the same steel cold worked, they are going to be sorely disappointed.

[Kevin R. Cashen]

O.K. here we go, this was almost done for a page at my website but I will share some here.

Forging is not a magical process and there are no secrets to beating the snot out of a piece of metal, but there are mechanisms and physical laws that it works by.

Plastic deformation of steel should occur one of two processes, slip or twinning, and more often slip. As previously mentioned steel does not like being compacted so you can’t just shrink it smaller with a hammer. The atomic lattice of the internal crystalline structure will not just implode upon itself because we wish it so. So since the those atoms just will not crush each other or occupy the same space at the same time, then something has got to give. Those atoms will slide around past each other however. But they must do it in a manner that obeys the laws of the universe, so they do it along particular planes within the lattice. These planes will be the ones that have the most atoms in them (stack a glass cube full of pin pong balls and then look at it from all different directions there will be one angle or plain that if sliced through the cube will hit the most ping pong balls) so these would be planes of higher atomic density. Face centered cubic structures (gamma iron, or steel heated above Ac1) have more planes coinciding with the desired slip plane {111}, so steel in the austenitic condition will deform much easier. Body centered cubic structures (alpha iron, or steel at room temperature) do not offer the same atomic arrangement since it is less dense with odd plains {110}, {112}, so alpha iron is resistant to slip and thus resistance to deformation. So you see it is the transition from high density to lower density that makes the whole thing possible!

Here is an image that I will be putting on my site that illustrates some of what I am saying:



The top image represents grains within the steel before deformation. The bottom image show the same grains after deformation and slip. Note the enlargement of the grain boundaries and how they get more "jagged". The enlargement on the right shows how atoms move across each other in order to accomodate this process.

To simplify this- try to squeeze a deck of cards by pushing straight down on them, nothing gives. Now stand them on end and push, still rigid. Now lay the deck flat and give a slight push from the side. Now the deck is a different shape than the original neat stack. You have just demonstrated slip. Now imagine a huge mass of hundreds of card decks all oriented in different ways and you will get an idea of how steel moves. The accumulative affect of plains of atoms slipping will actually be visible lines, at very precise angles, under a microscope.

Twinning is when the lattice distorts along a plane with a strange zigzag alignment, there is still no shrinking of the lattice but the atoms will sit off kilter to accommodate the move. Twinning is much more common in nonferrous metals with hexagonal configurations though, and I will be the first to admit that I don’t know as much as I should about nonferrous metallurgy.

In the process of slip there will be rows of atoms that just sort of end in the middle of the lattice, leaving and odd space. This is an edge dislocation and it can assist in slip, much the same way pushing a wrinkle across an area rug can move the whole rug eventually.

Now eventually, if enough of the rough irregular deformation occurs in steel that is below Ac1, the irregularities that can interfere with slippage will multiply to a point at which the metal will become very rigid and when the slip planes are pushed to their limits they will come part, and I don’t need to say what that means.

But all of these slip lines, dislocations and twins etc, will be obliterated as soon as the entire matrix realigns itself from BCC to FCC at Ac1.

I am sorry folks, but this is the way it works. I don’t say so, the fundamental laws of the universe say so, and all the hype full of our fantasizing and wishing won’t change any of it.

[Michael Stora]

Kevin, Your ability to explain these concepts in simple English is outstanding.

Mike

[T. Karlsson]

Nice of you Kevin to take your time and explain to us. Its really nice to get a good ide of how it works.

[Kevin R. Cashen]

Originally posted by Adrian Ko
Kevin, did I illustrate this accurately?

BTW, did you draw those illustrations? If so, your graphic art skills are catching up with mine!

I think you misunderstood a couple of my sentences, probably because of my poor punctuation.


1. I did not mean that the grain boudaries are enlarged by slip, I meant to say that you should note my image enlargement of the grain boundary area in which the interfaces are shown to be more more jagged.

2. The structure in not necessarily compromised due to slip it just has more stored energy from the strain and more inconsistancies in the lattice- vacancies, dislocations, twinning and the like. This can either be good or bad depending upon how you look at it. It will result in stiffer coldworked steel but it will all be undone on the next heat. To label it as "compromised" would be inaccurate.

3. So long as you keep the words "edge packing" anywhere in it please don't attach my name to this illustration in any way . It in no way describes the pseudo-scientific voodoo vomit that has been put in the corner of edge packing. This is an illustration of an actual, sound scientific principal of physical metallurgy. It is "plastic deformation" and it works in the real world, as opposed to any nonsense about packing molecules.

Sorry to be picky, but you did ask. I am a bit head strong when myth and fantasy is being held up against scientific principals.

[A. Ko]

Okay, I revised the illustration and the wording. Is this more accurate?