Carbon Steel Blades and Grain Size

Discussion in 'Knife Reviews & Testing' started by Juha Perttula, Jun 28, 2017.

  1. Juha Perttula

    Juha Perttula

    Mar 27, 2013
    Effect of grain size on impact toughness of carbon steel

    I have tested the effect of grain size on impact toughness of carbon steels. The impact toughness was measured by Charpy impact pendulum. The amount of energy absorbed by a specimen during fracture was measured and the result was divided by the cross sectional area of the specimen. Specimens were 3-5 mm thick steel slabs (un-notched). The grain size was determined with microscope. In Figure 1 is shown my test results for carbon steel at 60 HRC.
    it vs gs.png
    Fig. 1. The effect of grain size on impact toughness of carbon steel at 60 HRC.

    Grain size of 0.02 mm is normal situation. When a smith overheats a blade, he will get grain size around 0.05 mm. Grain size smaller than 0.01 mm can be attained when steel has carbides or other grain refining agents which prevent grain growth. Modern carbon steels usually have small (0.015-0.03%) addition of aluminium, which prevent grain growth. Al is not mandatory in steel standards, but it is common practice of steel factories to add Al. Figure 2 demonstrates the effect of Al addition on impact toughness. The steel with Al have smaller grain size and for that reason higher impact toughness. However, the difference diminishes when high tempering temperatures are used.

    it vs tt.png
    Fig 2. The effect of Al addition on impact toughness of carbon steel after different tempering temperatures. Steel with 0.025% aluminium (blue diamonds) is compare to a steel which does not have added aluminium (red squares)

    Also addition of about 0.15% vanadium prevents grain growth. Vanadium is easier to use and it have more reliable effect than aluminium, but it is more expensive. My blades are made of vanadinium containing 80CrV2 steel. They surely have small grain size, and for that reason, they have sufficient toughness at hardness level of 63 HRC.

    Why grain size affects on toughness ?

    The toughness decreases with increasing grain size, because coarse grain size promotes brittle grain boundary fracture. In the Figure 3 is shown the fracture surface of very fine grained and extremely coarse grained steels. The fracture surface of coarse grained steel reveals individuals grains.

    Fig. 3. fracture surface of fine (left) and extremely coarse (right) grained steels

    It is difficult to study fracture surface with light microscope, because its shallow depth of field. For that reason, I used electron microscope. I hardened three carbon steel specimens from different temperatures and got three different grain sizes. In Figure 4 is shown an elecron microscope image of fine grained fracture surface. The fracture is trans-granular; it propagates through the grains. Figure 5 shows fracture surface of semi-fine grained steel. In this case the fracture has mixed mode. Fracture goes along grain boundaries revealing some crystals, but it also goes through the grains. In Figure 6 is shown brittle grain boundary fracture of steel with large grain size.
    Fig 4. Trans-granular fracture of fine grained steel. Fracture propagates through grains.

    Fig 5. Mixed fracture mode of semi-fine grained steel. Fracture propagates along grain boundaries and through grains.

    Fig 6. Grain boundary fracture of corse grained steel.

    It is well known fact that fracture mode gradually changes from trans- to inter-granular with increasing grain size. But it is not fully understood why it happens. Anyway, grain boundary fracture is very brittle; therefore, coarse grained steel is brittle.

    Good tempering temperature for fine grained carbon steel is 180 C

    Tempering increases toughness in two different ways. It decreases hardness by making metal crystals more deformable, and it increases strength of grain boundaries. Thus, with increasing tempering temperature coarse grained steel can attain toughness of fine grained steel, as shown in Figure 2. But if you want high hardness you should use low tempering temperatures. Then make sure that you have fine grain size. According to my tests, blades with fine grained carbon steel have sufficient toughness after tempering at 180 C (350 F), which produces about 62 HRC hardness. But in many situations, even lower tempering temperatures can be used.
  2. me2


    Oct 11, 2003
    Excellent work. Oh to have access to an SEM.
  3. brj


    May 18, 2005
    Cool stuff (and nice site, btw).
    On a slightly unrelated note, why jera, sowulo, perthro?
  4. me2


    Oct 11, 2003
    What steel was used for the toughness tests? 80CrV2?
  5. Juha Perttula

    Juha Perttula

    Mar 27, 2013
    They are my runic initials. They look better than JSP and are easier to carve into wood. I am also attracted to their magical power.
  6. Juha Perttula

    Juha Perttula

    Mar 27, 2013
    The test results were for carbon steel about AISI 1080. 80CrV2 have similar toughness to AISI 1080 with Al addition.
  7. Twindog

    Twindog Gold Member Gold Member

    Apr 6, 2004
    Thank you for that interesting post.

    When we buy a knife, at best we're lucky to know the Rc hardness of the steel. But your first chart on grain size and toughness shows that Rc isn't the full story. Rc may tell us the strength of the steel, but not the toughness.

    From your website, you say that high alloy steels don't offer much of an improvement over properly heat treated carbon steels. Does that hold true for some of the better powder steels, like Vanadis 4 Extra, CPM M4 or 3V?
  8. Juha Perttula

    Juha Perttula

    Mar 27, 2013
    I planned serial knife blade production of a powder steel Vanadis 4, but in my tests it was not better than carbon steel. I was dissappointed. Then I changed my plan and developed the double hardened blade (edge 63 HRC; spine 50 HRC) and with Lauri we planned the final industrial hardening process. It have many different names: Lauri PT, progressive tempered, double hardened, edge hardened, zone hardened, etc.

    AISI D2 may have carbides in size of 0.01mm, the carbide size of the better powder steels, like Vanadis 4, is around 0.001 mm; therefore, the powder steel may have sharper edges. The thickness of very sharp edge is around 0.001 mm, or even thinner. Thus, it does not have room for many carbides, and the majority of the cutting work should be done by steel matrix. I have checked this with microscope. If you work with shaving sharpness, and I mean shaving a beard not hair, the carbides are not necessarily very useful.

    If you allow rounder slightly blunt edge, then the edge has more room for carbides, and then, no doubt, numerous amount of small carbides of powder steels will make a great job, as showed many times, for instance, with rope cutting tests.

    Carbides may improve abrasive wear resistance, but edge rolling or chipping can also spoil the sharpness, particularly, if you cut hard woods or bones. Hard blade resists edge rolling, and tough blade resists edge chipping. So, for me hardness and toughness are far more important than the carbides.
  9. Twindog

    Twindog Gold Member Gold Member

    Apr 6, 2004
    Thank you for that explanation. It seems to run counter to where the knife industry is heading -- to more complex, powdered alloys.

    I had presumed that diamond stones would sharpen the carbides as well as the steel matrix, but maybe that kind of carbide shaping takes unusual skill.

    Can you clarify as well that in your first graph, showing toughness of a blade with varying grain sizes in the steel, that all those points are blades with the same 60 Rc hardness? To me, that shows what most of us miss: that the micro structure of the steel is more important than the hardness.
  10. Juha Perttula

    Juha Perttula

    Mar 27, 2013
    Diamond stones sharpen the carbides, but if a carbide is not properly surrounded by a steel matrix it tend to drop of. For that reason, I think, the edge should not be thinner than the carbide diameter.

    Yes, in the first graph all specimens had 60 HRC hardness. Microstucture could be more important than hardness. Fortunately modern tool steels and stainless steels have alloying elements and carbides which prevent grain growth so that they generally are fine grained. But carbon steels have occasionally too large grain size. I have met some knives and other tools made from carbon steel which were broken due to large grain size.
  11. hugofeynman

    hugofeynman Gold Member Gold Member

    Jan 18, 2011
    Thank you for showing us your conclusions. Can't avoid to mention that others have reached this same conclusion years ago.
    Industry goes with people, towards the so called "super steels", full of carbides, very expensive so, in people's head (sometimes me included), should make better knives.
  12. Twindog

    Twindog Gold Member Gold Member

    Apr 6, 2004
    The powder steels are promoted as having much finer and better distributed carbides (at least compared to their ingot counterparts). So my thinking was that steels like Elmax and Vanadis 4 Extra and M390 would have plenty of steel matrix surrounding the carbides. And if we use very fine diamond abrasives, we could sharpen the carbides along with the steel matrix, while minimizing carbide tear out.
  13. me2


    Oct 11, 2003
    Juha, what testing led you to these conclusions? I recently finished some testing and reached the same conclusions. The difference in lower alloy steels (CTS BD1) and higher alloy steels (S110V) is not significant when used in hand held knives. I saw no clear difference when comparing the 2 in daily use or in controlled cutting after 600 feet of cardboard.
    Revolverrodger and hugofeynman like this.
  14. mete


    Jun 10, 2003
    Technically one doesn't prevent grain growth but slows it down .The grain boundaries must move as one grain takes over another [hostile corporate take over ].Also understand the other GB happenings such as collecting elements [good and bad ] . Much of the details lead use into 'micro-alloying !
  15. Blue Sky

    Blue Sky Gold Member Gold Member

    Jun 16, 2002
    Maybe a dumb question, but is it possible to determine grain size without a fracture? Perhap a highly polished surface and a high magification? Would be nice to have a way to non-destructively test a sample.
  16. Juha Perttula

    Juha Perttula

    Mar 27, 2013
    I have used leather cutting test. It is comparison test. Ten cuts, change knife, ten cuts, change knife, ... gradually winners and losers are found. Then knives are sharpened. Tests should be made several times.

    Here is test results of six steels with different hardness which I have studied most extensively
    Ultra high carbon steel with 1.6% carbon
    AISI 1080 carbon steel with 0.8% carbon
    AISI D2

    Winners: 1.6%C/67HRC and 0.8%C/66HRC
    Second: AISI D2 / 64HRC
    Thirds: 1.6%C/63HRC and 0.8%C/63HRC
    Loser: 1.6%C/60HRC

    This was very large and reliable test. The cutting was made by my friend who is leather professional. He did not know the steel grade or hardness. He just ranked the blades. In this test hardness was far more important than carbides.

    I myself have compared Vanadis 4 and AISI 1080 by leather cutting test, and did not found any differences.

    However, I think, people are too focused on abrasive cutting tests, because edge chipping and edge rolling may also spoil the sharpness.
    Last edited: Jul 5, 2017
    hugofeynman likes this.
  17. Juha Perttula

    Juha Perttula

    Mar 27, 2013
    To reveal grain size form polished surface you should etch the surface. With martensitic steels, picric acid with soap corrodes more grain boundaries than grains, but, in practice, this is the most difficult task in metallography. The picture is taken from the polished and etched surface by microscope. Grain boundaries of small and large grains are revealed.
    Last edited: Jul 2, 2017
  18. Juha Perttula

    Juha Perttula

    Mar 27, 2013
    Here is a picture how impact toughness is gradually decreased after demagnetization due to grain growth. This AISI 1080 carbon steel does not have aluminium, or other micro-alloying element, which retards grain growth. The hardness is 61 HRC

    Specimens were placed in a furnace at 800 C. Demagnetization occurred after about five minutes. Then first two specimens were quenched, after one minute, then one more specimen was quenched, and so on. Finally after 13 minutes the last two specimens were quenched.

    The specimen instantly after demagnetization is slightly "raw"; therefore, poor toughness. After one minute, grain size is still small and the toughness is good. (Notice that with forge hardening you can not wait one minute after demagnetization because the temperature of forge is much higher than 800 C) Between 1 and 5 minutes majority of grain growth takes place, and toughness decreases. From 5 to 13 minutes grain growth slows down, and toughness decreases only slightly.

    So, in unalloyed carbon steel majority of grain growth may take place withing 5 minutes. But with steel which have carbides, aluminium or vanadium micro-alloying, similar amount of grain growth may occur after heating of several hours or days. With overhigh temperatures grain growth is faster.

    I want to emphasis that this was laboratory test. With forge hardening of carbon steels you may not want to use one minute heating after demagnetization because the temperature of forge is much higher than 800 C and the forge fire easily overheats the steel. According to my experience it is nearly impossible to attain superfine grain size with forge hardening if a carbon steel does not contain large amount of carbides, or micro-alloying elements (V, Al). However, despite slightly coarsened grain size, toughness can be good after tempering around 270 C (tempering color purple to blue).
    Last edited: Jul 10, 2017
    hugofeynman likes this.
  19. me2


    Oct 11, 2003
    Can you elaborate on the "raw" condition?
  20. Juha Perttula

    Juha Perttula

    Mar 27, 2013
    "ripe" condition is austenite with homogeneous distribution of carbon. In "raw" condition carbon distribution was still uneven and perhaps there were still left some ferrite grains.

Share This Page