Thanks to Devon Craun for becoming a Knife Steel Nerds Patreon supporter! Your support is funding knife steel research.
Some materials like aluminum form a passive oxide layer that prevents further corrosion. Steel is not one of those materials. Instead, steel forms iron oxide, or rust, that doesn’t protect the underlying iron and flakes off leading to further corrosion. However, when sufficient chromium is added then a chromium oxide passive layer forms which protects the steel from corrosion in a similar way to a metal like aluminum with its own aluminum oxide layer. A simple schematic diagram shows the passive film vs rust [1]:
The cutoff point for how much chromium is required to be “stainless” is somewhat fuzzy and is sometimes given as 11 or 12%. However, a further increase in chromium improves the strength of the passive film. Here is a simple plot that shows the effect of chromium on corrosion in one environment [2]:
While molybdenum or nitrogen do not create a consistent passive film on their own in steel, they do strengthen the existing chromium oxide passive film, and the pitting resistance equivalent number, PREN, captures this effect:
PREN = Cr + 3.3Mo + 16N
I discussed molybdenum additions on corrosion resistance in stainless steels briefly in the article on 154CM.
Because D2 has ~11-12% chromium it is sometimes called a “semi-stainless” since it is assumed that with such a high chromium content that it must be very close to being stainless. However, because of its high carbon content (1.55%) a good portion of that chromium is tied up in carbides rather than being in solution where it can contribute to forming the passive film. The large fraction of carbides in D2 can be seen in its micrograph where carbides are the light grey or white particles:
With increasing austenitizing temperature (the high temperature steel is heated to prior to quenching) there is more dissolution of carbides and therefore more chromium in solution for improving corrosion resistance. This can be calculated with Thermodynamic software that estimates the amount of carbide and composition of the matrix at different temperatures. Here are calculations for D2:
Because of the high carbon and chromium content of D2, a certain fraction of chromium carbide is stable all the way up to the melting temperature. Therefore the chromium in solution maxes out at 9.3%, though that is at an unrealistically high austenitizing temperature where grain growth and poor toughness are likely. With an increase in chromium content, however, the chromium in solution can be increased. The higher chromium content also shifts the optimum austenitizing temperature, so I showed 1850°F (common temperature for D2) along with higher 1950 and 2050 °F temperatures which are in the range for stainless steels:
So to reach the somewhat arbitrary 11% chromium in solution, at least 16.6% Cr is required for 2050°F, and 17.5% Cr is required for 1950°F. The higher value of 17.6% Cr or so may be somewhat more realistic because 2050°F is a relatively high austenitizing temperature, higher than some furnaces can reach. Increasing Cr also decreases the carbon content in solution (which controls final hardness) and also increases the carbide content (higher wear resistance and poorer toughness). D2 with 16.6% Cr at 2050°F has similar carbon and carbide content but with the higher chromium in solution we are looking for. D2 with 17.6% Cr at 1950°F has somewhat higher carbide volume and lower carbon in solution, which would likely mean a reduction in hardness by 2-3 Rc:
A higher chromium stainless version of D2 is essentially Carpenter’s XHP steel which has 1.6%C and 16.0%Cr. The datasheet for XHP [3] says, “It can be considered either a high hardness 440C stainless steel or a corrosion-resistant D2 tool steel.” The compositions are shown below:
The chromium content is a little lower than the minimum 16.6% we calculated with Thermodynamic software, though the datasheet claims that XHP has “equivalent” corrosion resistance to 440C. They may be using a generous definition of “equivalent.” They also have data on toughness which appears to be similar to D2 [3][4], where izod unnotched impact toughness is approximately 27 ft-lbs at 60.5 Rc for both D2 and XHP. Abrasion resistance is also similar as shown below [3]:
Another question you may ask is how the corrosion resistance of D2 compares to other tool steels. Does its higher bulk chromium content make it better?
The steels are ranked according to Cr in solution. Obviously this is not an exhaustive list but hopefully gives a useful cross-section of tool steels. You can see that the PREN values are high for some of them due to high Mo content, though as discussed earlier in this article that likely doesn’t mean “stainless” levels of corrosion resistance as Mo enhances the existing Cr passive layer rather than replaces it. The ranking by Cr content in solution is probably more accurate for predicting the relative corrosion resistance of each. One thing that is perhaps surprising is that D2 is not the highest tool steel on the list; several 7-8% Cr steels are better including A8 mod, 3V, and Cru-Wear. This is due primarily to the lower carbon content though also vanadium additions to 3V and Cru-Wear means that some of the carbon is tied up in vanadium carbides and therefore not forming chromium carbides. So the status of D2 as being the high corrosion resistance “semi-stainless” is perhaps overblown. It has about half of the chromium in solution as stainless steels and there are several other tool steel options with better corrosion resistance. Actual corrosion resistance tests for most of these steels are not available, or at least not common, however, because they are not designed for corrosion resistance. Therefore we can’t easily compare the calculated chromium in solution vs corrosion tests.
One caveat to these calculations is that many of these tool steels are recommended to be heat treated with a high tempering temperature (>950°F). High tempering temperatures leads to some loss of corrosion resistance because very tiny chromium carbides are precipitated at high tempering temperatures (see my tempering article). That depletes the matrix of chromium and lowers corrosion resistance, as can be seen in the chart below [5]. Therefore to maximize corrosion resistance of both stainless and non-stainless tool steels lower tempering temperatures should be used.
[1] https://sassda.co.za/about-stainless/introduction-to-stainless-steel/
[2] http://emrtk.uni-miskolc.hu/projektek/adveng/home/kurzus/korsz_anyagtech/1_konzultacio_elemei/stainless_steel_case_study.htm
[3] https://cartech.ides.com/datasheet.aspx?i=102&E=343
[4] https://www.alphaknifesupply.com/Pictures/Info/Steel/D2-DS-Latrobe.pdf
[5] http://www.crucible.com/PDFs/DataSheets2010/Datasheet%20CPM%20S110Vv12010.pdf
Larrin,
Thanks for another great article. Your posts have been shedding light on many questions I have been pondering. Keep up the great work.
You must be pondering good questions!
as support for your table comparing pren for various toolsteels, i find k490 and k110/D2 to have about the same corrosion resistance will only rust if you place them wet in a sheath, low tempered both of them…
Thanks for a great read
Thanks for these articles Larrin. Your articles help me gain a better understanding and although I still feel I lack the comprehension on many of them, this one was a little easier to understand. It’s also nice to see some data behind some of the claims I’ve read.
Thanks for the great read!
Thanks!
Having studied some metallurgy for heat treating when I had to HT for a machine shop I never had a resource like Knife Steel Nerds, I had help from a HT company though. I wanted to say your website is a great resource and thank you for putting it out here Dr. Thomas. Although the internet was available, veridical heat treating metallurgy information wasn’t in large supply back then, 1998-2008. Downloading data sheets is what I did the most or emailing suppliers to ask “How, What, You sure?” LOL
I had been wondering about Carpenter’s stainless D2 XHP and I am trying to figure out its PREN as it isn’t on your equivalency list? Is it up near AEB-L or lower than that?
I knew some steels like CPM 3V tended towards being quite corrosion resistant as I’ve had woodcrafters act surprised to find out it wasn’t stainless when I told them to be careful near saltwater. I was really surprised to see D2’s 8 PREN being so low. I haven’t had that much of a problem with my own D2 knives is why I am so bewildered. I don’t use D2 anymore and only have two left that are over 8 years old I haven’t put handles on. I guess I should have sold them before D2 became a budget steel. lol
You say in the article that when you tested the cryo/non-cryo D2 in your 48 steels article that they were basically the same, but you tempered the cryo D2 at 500 to 59.2 Rc while the other you tempered at 400 to 60.5 Rc hardness which cut 13 less cards . I wonder what would the cryo D2 have been at a 400 temper? I’d guess around 61.5 and most certainly would have cut at least 50 more cards out of 500=10% more ballparking the math.
The difference in hardness 60.5 down to 59.2 is 1.3 points and equals a hardness difference of 6.5% using the Rockwell C Differential Comparator . So while being 6.5% softer the 59.2 D2 cut thirteen more cards. I guess you were aiming for the same hardness, but that isn’t what you got. What you ended up with I would not call ‘the same’. I suppose you’ve been so busy you missed the difference Dr. Thomas. The cryo treated D2 was actually better than the non-cryo blade that should have cut a significant amount more than the softer steel yet didn’t. I wonder if the cryo brought out any vanadium enhanced chromium carbides or does it ever in D2? There is certainly enough vanadium to make some vanadium carbides in D2, just look at CruForgeV.
For non-metallurgists out there (not that I’m one either), 60 to 61 Rockwell C scale is a 5% gain in hardness. Rc is exponential as you may know, so 50-51 is 4% so isn’t the same as 60-61. It is like a speedometer that’s off, the higher you go the more it differentiates. 60-62 is 10.3% not 10 for instance, but 50 to 52 is 8.2%.
I’ve tested the corrosion resistance of XHP twice, and found it to be quite low in corrosion resistance:
https://knifesteelnerds.com/2019/10/14/corrosion-resistance-testing/
https://knifesteelnerds.com/2021/03/25/cpm-magnacut/
There is no vanadium carbide in D2. You can’t compare with CruForgeV because it is the high chromium in D2 that leads to no VC.
That corrosion on XHP was terrible, they shouldn’t call it stainless at all.
Dr. Thomas does the vanadium in D2 enhance the chromium carbide at all? If it doesn’t why have the higher .85-1.00% and just leave it at .3% for grain refinement? I’ve seen some x-ray fluorescence testing of some terrible Chinese D2 way out of tolerances like .38% V and 10% Cr. To bad those PMI tests do not measure lighter elements like carbon or nitrogen as I’d like to see the carbon content of that steel. Wouldn’t be able to tell the differences of AEB-L from 14C28N without some serious cutting tests now I think on it.
The original patent didn’t give a reason for the vanadium. I imagine it would help with secondary hardening.