Science?

Joined
Jan 6, 2015
Messages
23
This is more of a scientific question. What is the actual scientific stuff that goes on in the blade when you heat treat it? Also with tempering?
 
Lol! It's called metallurgy, and it's pretty complex, look in the stickies, they give a pretty darn good over view of the subject, as it applies to knife making!
 
I was going to say that I just found that Verhoeven paper for free the other day. REALLY good stuff. If you read the parts about the two alternate ausquenching methods, I have a sneaking suspicion that you MAY be getting a brief glimpse into some of the "secret proprietary heat treating methods" we hear about. I am referring in particular to the second method that leaves you with like at least 82% fresh martensite, 13% lower bainite and the dreaded retained austenite at or below the magic 5% right out of the quench as best as I can tell. . What do you guys think? :D
 
What do you guys think? :D

I think it's all tremendously interesting... and also worth every penny to pay cats who've done it for a 1/4 century or more full-time to deal with it for me.
 
I have made some pretty long posts about exactly what is happening in the steel atoms and structures in the past. Some of that may be in the stickies. Other threads can be found using the BF search engine - https://www.google.com/cse/home?cx=011197018607028182644:qfobr3dlcra

Try searching using "Metallurgy Stacy" and just "Metallurgy"

For one of my funnier metallurgy treatises:
http://www.bladeforums.com/forums/showthread.php/842641-Blacksmith-Metallurgy-Explanation

This is a long post from a past thread that deals with the subject.

Steel is a combination of iron...usually about 99%, and carbon...usually about .6-1.0%.
To that they add small amounts of alloy ingredients. Some are to make it easier to make the steel, such as silicon, phosphorous, sulfur. Others are there to enhance the properties desired in the steel -Manganese, Chromium, Vanadium, Tungsten, and some others.
The alloy ingredients form various structures that are often in the form of carbides, which can make the steel harder and tougher. This comes at a price, though, over a simple steel alloy (with just iron, carbon, and a tad of manganese). Once you start adding things like chromium and vanadium, you need more heat and longer times for these to go into solution.
When the alloy becomes overstuffed with these ingredients, as in martensitic stainless steels, the iron content is reduced to as low as 70% and things like chromium, vanadium, and tungsten go up to as much as 20%. Carbon often reaches nearly the proportions of cast iron ( 2-3%). The excess carbon is needed because of the amount of carbides formed.

What does all this mean to knife makers??????
We want our blades to have a combination of two things - Hardness and Toughness.
This comes from creating a structure in the steel called Martensite. We can then manipulate the martensite into a desired blend of hardness and toughness.

So lets get out steel hot:
When we heat the steel up to a point where the atoms start to rearrange, the first major change comes around 1350F. That point is the Critical point - Ac1 or As, which means it is the beginning of the point where the steel converts to Austenite. Austenite forms at slightly different points depending on the alloy ingredients. The word "critical" means that there is a change in the physical properties (solid,liquid,gas, structure,etc.) at that point.
Next we reach the Curie point at exactly 1414F for steel. This point is the same for all steels. Its abbreviation is Tc. This is not a physical change, but deals with ferro-magnetics. At this point the steel becomes non-magnetic. Of all the "tricks of the trade" that smiths will tell you, this is the only one that is entirely accurate and repeatable. When the blade stops sticking to a magnet...it is at 1414F.
Then we reach the point where the alloy ingredients are able to go into solution. This is the target point. This point is often called Ac3 or Af. NOTE - Charts with the "s" and "f" used are easier to read, I consider that the letters mean "start' and "finish". For this reason, I will use those terms. The target is the place where all that we want to happen will, and things we don't want won't. Grain growth is the enemy when the steel gets too much above the target point.

Engineers have spent many years doing tests on steel to determine what it does and when it does it. Each steel type has its own set of data. This data is put into charts by engineers, because they do that sort of stuff for fun.
A look at one of these funny charts with funny names like TTT or ITC will show a bunch of curved lines. The curve sort of looks like a nose, and the left-most spot is called "The Nose". That "Nose" is usually about 1000F on the chart. This is the point where if your steel goes any amount right of the nose in its cooling ( takes too long to cool down), it will become partly or all pearlite ( which is not what we want). To determine what you want the chart to tell you, look at the base of the chart and the vertical side of the chart. Along the base you will see the time experssed in seconds, and then in minutes. On the vertical side you see the temperature scale.. Thus the chart shows how soon things happen at any specific temperature . The X/Y coordinates of any spot on the curved nose line will tell you how much time you have to get down to that temperature during quench.
All those funny letters and names assigned to the places along the temperature chart tell you where the steel changes from one structure to another. A= austenite; M= martensite; Ferrite is iron and carbon ( body-centric for those engineer types); Cementite is a hard and brittle structure of iron and carbon; Pearlite is a soft structure made up of layers of ferrite and cementite.
Lower case sub letters, "s and f" mean start and finish. Letters "c and r" mean climbing and returning...or simply heating up and cooling down. Note that things don't always happen at the same point rising and falling.

What we are concerned with is taking a piece of steel that we have shaped into a blade and changing it from one structure to another in a controlled process we call "Heat Treatment".
We do this by first heating the steel up in an environment that won't ruin the blade. This is part the atmosphere of the forge, and part how we protect the blade form elements we don't want to add to our steel....mainly oxygen. That is for another discussion, though, and for this talk we will assume your forge/oven atmosphere is right and you did the proper things to protect the blade form evil oxygen.

Back to the funny chart. The steel starts off a mix of God-knows-what structures after we have been working on it. In the final Heat Treatment we heat it up and as it passes As/Ac1 it starts to change into austenite. Then it passes the Curie point, and our trusty magnet ceases to attract the blade. On we go as we heat up the blade until we reach the Af/Ac1 spot. This is where we want to stop heating the blade and hold it at that point ( as close to evenly as possible) during what is called the "soak". While the blade is "soaking" the alloy ingredients have time to slowly recombine into various structures, go into solution, and the evenly distribute themselves. This takes from 5 to 10 minutes for simple steels, and up to an hour for complex stainless steels. As a general rule of thumb, a 10 minute soak is used for carbon steel alloys, and 30-45 minutes used for stainless steels.
The temperature of this varies depending on the alloy ingredients. Simple carbon alloys are usually about 1400F to 1500F, and stainless steels are around 1850F to 2050F.

Once we have allowed the ingredients to go into their solutions and such, we now need to cool the blade down and make Martensite out of it. This is the "Quench" stage. By rapidly dropping the temperature past the "Pearlite Nose" we can super-cool the austenite and thus not make any pearlite. Notice that once past that nose, the amount of time starts to expand as the blade cools down. This is why we need different quenchants and methods for different steels. All we want to do is get past the nose quickly, after that we want to cool down at a rate that allows the steel to convert to martensite , but does not shock the steel any more than necessary. For some steels, like the simple 10XX series, this is fast, one second to pass the nose, but as we add those alloy ingredients, the amount of time to pass the nose increases. The "O" series steels have about ten seconds, stainless steels give you many minutes to pass the nose. Thus, the proper quenchant for these steels would be -
Water or fast oil for 10XX steels
Slower oil for O series and alloy steels like 5160
Air for D-2 and stainless steels.

Once past the nose, we can slow down the cooling rate. The steel needs time to prepare for the sudden change that will occur at the point where it changes to martensite. At this point we still have austenite ( hopefully) and the steel is rubbery soft and in no danger of cracking or breaking. As we reach the Ms point, around 450-500F, the steel rapidly converts into martensite. This is where the cracks come from. The change in structure can tear the blade in half. Thus we need to cool the blade across this area as slowly and evenly as possible. The choice of quenchant can make or break ( literally) a blade here. At about 200F the austenite has mainly converted to martensite. Any left over austenite is called RA, or "retained austenite". This is a very small amount in simple steels, and is dealt with in the tempering stages. However, with the complex stainless steel alloys, this RA is a big problem unless converted as fully as possible to martensite. The much larger percentage of RA at room temperature is due to the fact that ,while the simple steels finished their conversion to martensite (Mf) at about 200F, the stainless steels need to drop to about minus 200F, yes 200F below zero, to complete the transformation. This is done by adding a step called "cryo". The steel can be very unstable with part of it being very brittle martensite and part very rubbery austenite, so it is wise to give it a short stabilizing step called a "Snap Temper". This is a short temper cycle of about one hour at a low temperature of about 200F. This will stabilize the martensite somewhat and can avoid cracks forming in the cryo stage. After this "snap temper" the blade is cooled to room temperature and them placed in a chamber that will cool it to the sub-zero point where the Mf will finally be reached. This is typically done in liquid nitrogen, which is a bit below -300F, or in a solution of acetone/alcohol/kerosene and dry ice, which will cool the blade to about -100F. This is low enough to get most of the austenite converted, but the LN is certainly better. Economics and frequency of use usually decide which type of cyro one uses.
After the austenite is converted, it is a very strained and brittle structure. This is called Un-tempered martensite. This is corrected by tempering the steel at points between 350F and 1050F. The tempering target point is determined by a tempering chart which those engineers have also made up for you. The chart is much easier to read. It shows the estimated hardness at any given temperature on the curve. The first temper cycle ( after the snap temper) is to convert the martensite to tempered martensite, which is tougher, as well as convert any remaining stray bits of austenite into martensite. This newly converted martensite will, however, be un-tempered, so you need to do a second temper after cooling to room temperature. A second cryo treatment between the tempers will accomplish very little, and this is not usually done. To make it clear....cryo is done as a continuation of the cooling process during quench. After that cooling has been stopped and tempered, any cryo will be much less effective.

To state it in a flow chart for stainless steels:
Heat to austenitization > Soak > Quench > Snap temper > Cryo > First temper > Cool to room temp > Second temper
 
Stacy, that's a wonderful summary, thank you for taking the time to write it.
Can you please explain what the distinction is between pearlite and retained austenite? In practice they seem similar, in that they are much more flexible than we want in a blade.
I understand that pearlite is ferrite/cementite, but since it behaves as RA would, it's kind of a loose thread in my understanding- thank you!
 
Pearlite is a different atomic structure from austenite.

Pearlite is a lamellar ( layered) structure of iron ( ferrite) and iron carbides ( cementite). It is soft because the layers are easily shifted/sheared. For grinding it, the carbides are best balled up into spheres when annealing. This is called spheroidizing, or spheroidally annealed.
Pearlite is soft and tough.

Austenite is a Face Centered Cubic ( FCC) structure of carbon and iron atoms. The alloying ingredients and carbides are distributed along the boundaries of the cubes.
Austenite is a little harder than pearlite, and is very tough.

Retained austenite is the amount of austenite that did not convert to martensite in the hardening and tempering process.

Martensite is a body centered cubic (BCC) arrangement of carbon and iron atoms that forms when the austenite is super-cooled below the pearlite range. Once the austenite reaches the martensite range, it converts into martensite. With the right amount of carbon and iron and little alloying, this is a nearly 100% conversion. With higher alloying, the conversion may be less.
Martensite is hard and brittle. It needs tempering to make it less brittle and tougher.

Some RA is not a bad thing. It slightly toughens the blade against breakage and wear. However, it lowers the total hardness a bit. The tradeoff is very small for any of the steels we normally use for knives, and shouldn't keep anyone up at night except theoretical metallurgists and razor blade engineers.

The answer to not worrying about RA and such is a proper HT with known steel.
If the numbers are right, the math will almost always come out right.
If the numbers are right, you don't even have to do the math.
If you don't know some of the numbers, and are guessing at most of the rest...well, the answer probably isn't worth squat.
 
First, I'd like to say thank you! Thank you for asking questions and wanting to learn.

Next, Stacy is dead on. Known steel with a known heat treat is the BEST way to make the BEST knife possible. Yeah, I know your grandpappy made a knife once from a lawn mower blade forge welded to an old planer blade forge welded to the leaf spring off his old pickup that could skin three deer, one moose, and something that appeared to be a ferret and still shave hair afterward. However, anecdotal stories don't trump science. Confirmation bias is still bias.

Take a look at some of Kevin Cashen's site. Pretty good stuff from a really smart dude.
http://www.cashenblades.com/metallurgy.html
 

If the numbers are right, the math will almost always come out right.
If the numbers are right, you don't even have to do the math.
If you don't know some of the numbers, and are guessing at most of the rest...well, the answer probably isn't worth squat.

I'm making this a poster and hanging it in my shop!
 
I'll expand on the pearlite thing.
Tempered martensite is when you make the very brittle martensite a bit like pearlite. This softens the martensite and makes it tougher. The higher the tempering temperature, the more "pearlite like" the martensite becomes. Long enough and hot enough will basically make it back into pearlite.

Where most folks get this wrong is at what point this starts to happen. People want to temper their blades at 300F, 325F, and 350F. That isn't doing much of anything to improver the blades toughness, and leaves the blade too brittle for most tasks. Sure it is hard as woodpecker lips and still at near "at quench" hardness, but a temper of 400F or even 450F will greatly increase the toughness while only dropping the hardness a small amount. You have to go over 500F before the hardness drops significantly enough to affect a knife's abilities.

Sharpness has nothing to do with hardness, so an Rc 64 blade and an Rc 62 blade and an Rc 59 blade can all attain the same sharpness ( sharpness is due to geometry mostly). But, an Rc 64 blade may chip easily, and an Rc 62 blade may be hard to sharpen, and an RC 59 blade may be a great performer and easily touched up to keep it shaving sharp. The difference in temper on these three for a carbon steel is usually 350F, 400F, and 450F.

Don't fool yourself into thinking that making the blade harder will make the blade better. To be honest, Rc numbers are only useful for fine tuning a HT regimen. Once you get reliable HT results and methods, the actual hardness of the blades will vary little at all. Checking every now and then is nice, but with good procedures, the results will usually be good. People who are making blades for fun and such shouldn't spend any time worrying about the final RC hardness. Getting a good flat bevel, a proper edge thickness, a good HT, and proper sharpening techniques will matter far more than if it is Rc62 or Rc58. In out grandfather's days, almost all knives were in the low to mid Rockwell 50's. No one ever saw a problem with that because they were well made from good carbon steel, with good geometry .... and everyone knew how to sharpen knives right.
 
Sharpness is a matter of thin edge and the right geometry. An unhardened knife will get as sharp as a hardened one. It just won't stay sharp for long. Paper and tin can lids can be very sharp.

While the proper hardness is important on a knife blade, it doesn't determine the sharpness....and certainly not within the difference between Rc59 and Rc64. Steel choice can determine it to some degree, as some steels can attain finer grain than others.

There are three things that the steel type and HT will determine.
Hardness
Wear Resistance
Toughness
All three together determine how well a blade performs. On the first cut three blades sharpened identically would all cut the same. On successive cuts higher hardness would increase wear resistance, but decrease toughness. Just the right amount of toughness would slow down the edge wear, and increase the edge life. Too much toughness will lower the hardness and make the blade dull too soon. Changing the HT will make toughness and wear shift to allow more cuts. The perfect blend is a mating of enough hardness to allow a good edge life ( wear resistance) and have just the right amount of toughness to resist micro-chipping. For most general steels and use knives, that falls around Rc 58-59. Thin slicers can go harder, and rough use choppers lower. Carefully selecting the steel to gain more toughness, better wearability, or higher hardness is half of the way to make the most of a specific blade, and balancing the HT is the other half.

I can't find what you are referring to in that link?
 
http://www.tameshigiri.ca/2014/01/11/razor-edged-2-hardness-structures-and-sharpness/ Stacy, I think sharpness is very much dependent on hardness. Isn't that you in the picture of that article?

6b615e10469d20cfeb721cf3fd1fd0f5_zpsbe1cad7d.jpg
 
No, that is J. McDonald. He has been pounding on that same blade for three years now. Looks like he set his beard on fire.

(very few will catch this joke)
 
Back
Top