Dry Ice vs Liquid Nitrogen vs Freezer – How Cold Do You Need for Cryo?

Thanks to Gerog Belnap, Mingjim Yang, Jay Gonsalves, Reggie Wheeler, Charles Drouin, Daniel Bingham, Charles Platt, and Adam Andreasen for becoming Knife Steel Nerds Patreon supporters!

Video Version

Here is the video version of the following content:

Cold Treatments of the Past

I have a series of three articles about cryo where I introduced a good amount of background material about how it works:

Cryo Part 1: Maximizing Hardness

Cryo Part 2: Toughness and Strength

Cryo Part 3: Wear Resistance and Edge Retention

I also have a chapter in my book about cryogenic processing. I recommend the book to anyone who wants to learn about knife steel and heat treating, of course.

Cold Treatments and Cryo Explained

Some time ago I did some experiments with AEB-L where I compared the freezer and liquid nitrogen for AEB-L and found the following:

Using my household freezer (not a fancy freezer) led to a result that was surprisingly different than room temperature quenching alone. Common knowledge on the internet from knifemakers says that a freezer doesn’t “do anything” and that you need to have at least dry ice to have a true cold treatment for your knives. This shows a fundamental lack of understanding about what happens in cold treatments. When you quench the hot steel you are transforming from the high temperature phase, austenite, to the high hardness phase martensite. Martensite is what gives knife steel its high hardness, strength, and wear resistance. The transformation to martensite is primarily controlled by temperature, not time. So to get more martensite you need to quench below the martensite start temperatures (Ms) and as it cools down it forms more and more martensite as it approaches the martensite finish temperature (Mf) where the steel is fully martensite. If the steel does not reach the martensite finish temperature then there is some amount of austenite that remains in the steel untransformed, which is called “retained austenite.” The martensite start and finish temperatures are most greatly controlled by carbon; the more carbon is “in solution” in the austenite the lower these transformation temperatures are:

Data adapted from [1]

Other alloying elements also reduce Ms and Mf such as chromium. The higher the austenitizing temperature for a given steel the more carbide is dissolved which puts more carbon and alloy in solution for reduced martensite transformation temperature. Read about austenitizing in these articles: Part 1 and Part 2.  Therefore, austenitizing higher potentially means more retained austenite because Mf can be reduced below room temperature. At the simplest level, retained austenite leads to reduced hardness, which is why the AEB-L quenched to room temperature peaked in hardness from 1925°F (~62 Rc) and then dropped with increasing austenitizing temperature, down to 59 Rc from 2000°F. Small amounts of retained austenite can lead to an improvement in toughness so it isn’t always clear whether it is desirable to further reduce retained austenite. Virtually everyone agrees, however, that beyond about 15-20% retained austenite is undesirable as it leads to reduced strength and often reduced toughness. And luckily 15-20% RA is usually found at about the peak hardness. See below for Caldie tool steel as an example where they used different austenitizing temperatures and tempered to the same approximate hardness. In a compression test they found that the “yield strength” was reduced even with the same hardness with more RA. This means that the steel deforms more easily such as edge rolling even though the hardness test would lead you to think that strength would be sufficient.

Data adapted from [2]

Another study I have talked about is an old one where they looked at different cold temperatures after quenching with T1 high speed steel. They tried different periods of delay between the quench to room temperature and then putting it in the cold bath. They found that retained austenite stabilizes while the steel sits at room temperature, and the longer you wait before cooling the steel further, the less retained austenite will transform during the cold treatment.

Image from [3]

A cryo treatment such as with liquid nitrogen (-310°F) reduced the retained austenite from 20% down to 8% when placing it in liquid nitrogen directly after quenching. This degree of transformation was basically the same even after holding the steel at room temperature for an hour prior to cryo. Beyond an hour of waiting the effect became less and less until the maximum time of 20 hours where the liquid nitrogen was having about half the effect as when going directly into liquid nitrogen. Dry ice temperatures (-100°F) had nearly the same effect as liquid nitrogen when cooling directly after quenching, but after waiting for 20 hours it transformed almost none of the retained austenite. Freezer temperatures (~0°F), had about half the effect of liquid nitrogen if going into the freezer immediately, but waiting only a few minutes led to no change in retained austenite.

Similar trends were seen with W1 tool steel which is a simple carbon steel similar to 1095. They austenitized the steel from 1450°F and 1550°F to see the differences with austenitizing temperature:

Images from [3]

With a room temperature quench from 1450°F, the steel had 5.8% retained austenite, and from 1550°F it had 12% retained austenite. This is from more carbon being in solution leading to reduced Ms/Mf. Using liquid nitrogen temperatures effectively reduced the RA to near 0, as long as the steel was cooled immediately after quenching. Following similar trends from T1, the colder the temperature the less a delay affected the cold treatment.

New Cold Treatment Studies

Recently I ordered some steaks online and the box came with dry ice in it. I wanted to test it in comparison with liquid nitrogen and the freezer such as I had done with AEB-L before. I repeated the full experiment using 1900-2025°F in 25°F increments.

There were a handful of surprises in this data. One is that dry ice looked more similar to the freezer than it did to liquid nitrogen. Perhaps the lower martensite start temperature in the AEB-L led to more separation between dry ice and liquid nitrogen. Another surprise was that this steel I tested required about 50°F higher austenitizing temperatures for the same hardness as the original steel I tested. I realized that this was because the “AEB-L” was produced by Buderus rather than Uddeholm. One of the major knife steel suppliers sells Buderus X65Cr13 as AEB-L because Buderus and Uddeholm have the same parent company and the steels have the same composition. However, there is a microstructure difference between Buderus X65Cr13 and Uddeholm AEB-L which leads to a difference in hardening response. The knife steel supplier also sells the X65Cr13 to several other knife steel suppliers as AEB-L; you need to know which you have purchased to know how to heat treat it.

Buderus X65Cr13

Uddeholm AEB-L

Tempered Steel

After tempering at 300°F or 350°F, the hardness trends were somewhat different:

The peak hardness for room temperature quenched steel was raised from 1950°F up to 1975-2000°F after tempering. The hardness was increased by “precipitation strengthing” which can be seen at either low tempering temperatures (~100-300°F) or high tempering temperatures in the case of high alloy steels with Cr, Mo, W, V, etc. (~900-1100°F). I stated earlier in the article that the steel shouldn’t be austenitized above the peak hardness temperature, but this becomes somewhat difficult when the peak was 1950°F without tempering but 1975-2000°F after tempering. This is a question that needs further exploration. Perhaps to be on the safe side I would stick with 1950°F.

Toughness Testing

I used 1925°F as the temperature from which to do toughness coupons, which I tempered at 350°F:

The hardness was increased by using the cold treatments. The freezer specimens ended up slightly harder than the dry ice specimens, just due to the natural variation between coupons. The liquid nitrogen led to the highest hardness as expected. The toughness was reduced with the increase in hardness and drop in retained austenite. However, the drop in toughness was relatively small for the amount that hardness increased, so in general I would say that the liquid nitrogen processing was worth it in this case. I also compared these toughness coupons to earlier Uddeholm AEB-L and Buderus Nitro-V. The toughness was similar to Buderus Nitro-V (a very similar composition), both slightly reduced from Uddeholm AEB-L. This is likely due to the microstructure difference between the two products.

In a study on Z-Wear/CPM-CruWear we found an increase in hardness with cryo but any change in toughness was within the statistical “noise” of the test:

In a study on 52100 we found a small decrease in toughness even after compensating for hardness:

I have seen a case or two where a significant drop in toughness was found with cold treatments. There was a study on Elmax steel where the found the toughness was dropped by more than half. I haven’t seen anything this extreme in my own testing so I don’t know if this is due to the austenitizing temperature, something specific to Elmax, or a problem in the testing of the researchers:

Image from [4]

There is definitely more research that could be done in this area. The difference in toughness with cryo on AEB-L would definitely be different with another austenitizing temperature because more retained austenite would be in the room temperature condition. And different steels may respond differently as well.

Wear Resistance and Edge Wear

I summarized the journal article studies on wear resistance changes from cryo in the Part 3 article that I linked to previously. Since that article I did a large CATRA study, and part of that study I did a cryo comparison for edge wear. There is a commonly cited study on D2 where they did cryogenic processing for different lengths of time and found that 36 hours was the optimal cryo time for maximizing edge retention. However, when I plotted the different wear tests they did against the measured hardness, the 36 hour cryo condition looked more like an outlier; all of the tests have a fairly simple trend with hardness apart from the 36 hour condition.

Data summary from journal article [5]

Journal article data summarized by me

So for my CATRA test I did two heat treatments, one which had no cryo and was tempered at 400°F, and one which had a 36 hour cryo and was tempered at 500°F. I did the two different tempering temperatures in an attempt to get the hardness of the two conditions the same, though the version that had no cryo actually ended up being slightly harder. Because of the higher hardness, the knife with no cryo ended up having higher edge retention. Whatever mechanisms the researchers thought would show superior wear resistance with a 36 hour cryo didn’t have enough effect to overcome a small hardness disadvantage. Therefore I think cryo has the potential to increase edge retention from wear from increased hardness, but we should question claims that other mechanisms are in play that can increase wear resistance in other ways.

Frequently Asked Questions

Which steels “need” cryo?

With virtually any steel, as long as you are using an austenitizing temperature at peak hardness or below, the steel is going to be fine. Beyond that there is excess retained austenite which reduces strength. This limits the hardness for some steels more than others. LC200N and Vanax, for example, top out around 60-61 Rc even with cryo, and a couple points less without it. Higher hardness is possible with cryo, of course, and in some cases you get an increase in hardness with little change in toughness as pointed out in the studies above.

Does cryo do anything to carbon steels and low alloy steels?

As shown earlier there was a hardness and toughness difference with 52100. And of course the study on W1 in the literature with different cold treatments. It changes things. Here is the full hardness chart on 52100 that we measured between using liquid nitrogen and without:

How long should the cryo treatment be?

Martensite forms basically instantly so it only needs to reach the temperature of the freezer, cold bath, etc. Some studies have recommended extended times for increased wear resistance but those studies are questionable. Usually 60 minutes is fine for cryo or cold treatment.

Should you temper before a cold treatment?

Tempering before a cold treatment reduces the chance of warping or cracking. However, it also stabilizes retained austenite and makes the cold treatment less effective, just like allowing the steel to sit at room temperature. Therefore I don’t recommend doing a temper beforehand. Think of cryo as an extension of the quench down to lower temperatures.

---

[1] Kobasko, Nikolai. “An explanation of possible Damascus steel manufacturing based on duration of transient nucleate boiling process.” In Proceedings of the 8th WSEAS international conference on fluid mechanics, 8th WSEAS international conference on Heat and mass transfer, pp. 81-86. World Scientific and Engineering Academy and Society (WSEAS), 2011.

[2] Rehan, Muhammad Arbab, Anna Medvedeva, Berne Högman, Lars‐Erik Svensson, and Leif Karlsson. “Effect of Austenitization and Tempering on the Microstructure and Mechanical Properties of a 5 wt% Cr Cold Work Tool Steel.” steel research international 87, no. 12 (2016): 1609-1618.

[3] Lement, Bernard S. Distortion in tool steels. American Society for Metals, 1959.

[4] Ducki, Kazimierz J., Jakub Jasiewicz, Grzegorz Junak, and Lilianna Wojtynek. “Effect of heat treatment on the microstructure and mechanical properties of sintered stainless tool steel.” Solid State Phenomena 226 (2014).

[5] Das, D., A. K. Dutta, and K. K. Ray. “Optimization of the duration of cryogenic processing to maximize wear resistance of AISI D2 steel.” Cryogenics 49, no. 5 (2009): 176-184.