The Secret Heat Treatment of Frank J. Richtig

Thanks to Dean Baughman, Kitoc420, Steve Callari, Jay Ghoo, and Mitch Cagile for becoming Knife Steel Nerds Patreon supporters!

Legendary Heat Treatments

Last week I wrote about What a Good Heat Treatment Can and Cannot Do, and as part of that topic I wrote about how some knifemakers have a legendary or even mythical reputation for their heat treatments. In that article I argued that the big differences are between “bad” and “good” heat treatments, and that the differences between various good heat treatments are much smaller. And that edge geometry and knife design are more important to knife performance than the differences that are possible between different “good” heat treatments. So I think it makes sense to discuss a particular case of a knifemaker known for legendary, unmatched heat treatments, which brings me to…

Frank R. Richtig: Knifemaker, Showman

Frank J. Richtig became famous for using his knives to chop through pieces of steel – bolts, axles, etc. and then slicing paper with the edge immediately after. It is said that the reason his knives were capable of such feats was his secret heat treatment developed in the 1920’s and early 1930’s [1]. He gained further notoriety when he was featured in Ripley’s Believe it or Not in 1936.

Image from [1]

Image from [2]

Research Published in a Major Materials Journal

Because of this interesting background to Richtig and his impressive knives, researchers at Lawrence Livermore National Laboratory in California investigated Richtig’s knives, which resulted in a paper published in 2000 in Materials Characterization [2]. They were intrigued by the unknown heat treatment he used, where they quoted Glen Lambert who visited Richtig in the 1960’s. Lambert wrote an article on Richtig that appeared in Knives annual 1984. When Lambert asked how he heat treated his knives, Richtig replied “A man is entitled to some secrets and that’s mine,” and “[He] told me that he had found no one worth of being told or taught his knifemaking, and until he did, nobody was going to get his ‘secret.’ During his declining years, no one came forward, and when he passed away in the early 1970s, so did the tempering process.”

The researchers looked at two knives, one 180 mm (7-1/16″) long and the other 100 mm (3-15/16″).

Image from [2]

They measured the composition with spectroscopy and found the steel to be 1090 or 1095, with 0.86-0.93% carbon with small amounts of Cr, Mn, and Si. So the steel used was nothing out of the ordinary in Richtig’s knives. They also measured hardness for the knives with the large knife ranging between 36-46 Rc with an average of about 39 Rc and the small knife 40-56 Rc with an average of 50 Rc.

The researchers also looked at the microstructure of the knives which they interpreted as showing “a uniform distribution of spherical, proeutectoid carbide particles in a matrix of ferrite and fine cementite (white particles within the matrix).” It is difficult to differentiate between bainite and martensite, so in comparisons with 52100 in both microstructures they hedged their bets, “the microstructure of the Richtig steels matches most closely the microstructure of the 52100 steel Q&T at 500°C or austempered at 400°C.” I previously wrote an article about the differences between bainite and martensite, which you should read to follow the rest of this article.

Large knife low magnification [2]

Large knife high magnification [2]

Small knife low magnification [2]

Small knife high magnification [2]

In an attempt to determine if Richtig’s knives were made up of bainite or martensite, they performed tensile tests on the large knife steel (which was 39 Rc on average). They used very small tensile specimens because they had to take the tensiles out of a knife rather than a large bar of steel. Tensile testing involves pulling a piece of steel until it breaks. This results in a “stress-strain” curve which represents the properties of the steel. Strain is a measure of how far the steel was pulled, and stress is a measure of the load applied to the specimen. This results in a few important properties such as the “yield strength” which is the stress at which the steel has begun to permanently deform (as opposed to returning to its original shape), the total elongation (how far the steel was stretched prior to breaking), and the ultimate tensile strength (the stress required to break the piece). I introduced tensile testing in this article on flexing steel, and how it relates to toughness in this article.

They also compared the measured tensile properties of the Richtig knife with various heat treatments of 52100 to determine which was closest to Richtig’s knife. They first compared with “quench and tempered” 52100. Quenched and tempered steel is the most common method of heat treating knives where the steel is heated to high temperature (austenitizing) and then quenched such as in oil or water, and then tempered at a lower temperature such as 200°C (400°F). Comparing with quenched and tempered 52100, they found that a 500°C temper of 52100 led to similar ultimate tensile strength (I’ve marked in blue), but the yield strength (marked with red) was much lower with the Richtig knife. So they concluded that the Richtig knife was not quenched and tempered.

They next compared with “austempered” 52100. Austempering is to form bainite, where instead of quenching to room temperature in oil, it is quenched to an intermediate temperature such as 250°C (500°F) and held there to form bainite. Tempering afterward is optional as austempering is sort of like quenching and tempering in one step. They found that the yield stress was significantly lower in the austempered 52100 (as opposed to Q&T) so they concluded that Richtig likely austempered his blades, which would help explain the low yield stress. So they proposed that Richtig heat treated his knives by austempering at approximately 400°C for the large knife and 300°C for the small knife, the different austempering temperatures explaining the difference in hardness between the blades.

The researchers wrapped up their study with the following comment:

“The authors did successfully cut a 6-mm diameter mild-steel bar by hammering the Richtig blade through it, without damage to the blade. An attempt to duplicate this experiment with a contemporary, high-carbon kitchen knife caused the knife blade to chip.”

Follow-Up Study on Recreating Richtig’s Heat Treatment

Two senior undergraduate students, under the supervision of professors, attempted to replicated Richtig’s heat treatments using the proposed austempering heat treatment procedure [3]. They used 1095 steel of a similar composition to the steel used by Richtig. Because they did not know the exact heat treatment used by Richtig, they used a range of austenitizing (760-900°C) and austempering temperatures (350-450°C). They used full size ASTM tensile specimens. Strength varied over a wide range, including values close to those measured from the Richtig knife. However, a combination of ultimate tensile strength and total elongation matching Richtig (1220 MPa and 14.5% TE) was not achieved. Toughness of steel is roughly the energy under the stress-strain curve, so higher toughness is obtained by having a high combination of strength and ductility (elongation). Therefore they also apparently did not match the toughness of the heat treated steel in the Richtig knife.

The researchers concluded that they did not match the ductility of Richtig’s knives “due to processing defects during the manufacture of the steel purchased for the study” after noting carbide bands in the steel they used. Segregation of alloying elements and carbide bands can reduce ductility.

Carbide bands in the 1095 steel used in the study

The researchers looked at the microstructure of the steel they heat treated and found that using sufficiently high austenitizing temperature along with austempering would lead to a bainitic microstructure free of pearlite and ferrite. I have pointed out the bainitic microstructure which has a lath-like appearance which has very tiny carbides making it look “fuzzy.” I’ve also pointed out the larger “proeutectoid” carbides which are those not dissolved during austenitizing.

Image from [3]

However, with the low austenitizing temperature, such as 760°C, the steel was not fully austenitized, leading to ferrite and pearlite in the microstructure:

My Richtig Knife

I was able to find an inexpensive Richtig knife on eBay recently. It may have been so cheap because a small piece of the tip was broken off, which doesn’t affect why I am buying it: to study it. It is a butcher knife, which appears to have been one of his common models. It is about 0.055″ at the spine and 0.012-0.14″ at the edge. The grind is very consistent, and the handle relatively simple and comfortable. It has a cast aluminum handle like the majority of his knives did. His back handle stamp is a simplified illustration of his picture in Ripley’s Believe it or Not, which is fun. So far I have only measured the hardness of my knife, but will eventually cut it up to do microscopy and microhardness.

Critical Analysis of the Published Articles

In the first paper from the Lawrence Livermore researchers they stated they could not cut through steel with a modern kitchen knife unlike with the Richtig knife. And in the second paper they could not match the strength-ductility combination (toughness) of the Richtig knife. Between those two factors it can appear that Richtig indeed had a legendary heat treatment which is difficult to match even with modern methods. However, there are several factors to keep in mind.

Inconsistent Heat Treatment

The two knives measured for hardness had a wide variation, both within a single knife and between the two knives. My Richtig knife measured at 57 Rc, higher than either of the two they tested. This wide variation in hardness tells me that Richtig’s heat treatment was not very consistent, which would point to the heat treatment being unimportant to its apparent good performance. With the attempted replication of the Richtig tensile properties, only one knife was known so that is the one they tried to replicate. It is not known how representative that knife is. The 57 Rc knife would have very different properties. In the Materials Characterization article they proposed that the two knives had different hardness because of the selection of a different austempering temperature (300 vs 400°C). However, that would not fully explain the wide variation of hardness within a single knife. And I don’t think he was austempering at all as I will explain further in this article.

Gage Length

The two papers used two different tensile specimens. This may not seem important on the surface, but the “gage length” (shown below) has a strong effect on the recorded total elongation. This may explain why the total elongation was too low in the tensile specimens used with the attempted replication. The Lawrence Livermore researchers used a tensile specimens similar to “E” below with a 12.5 mm gage length while the researchers who attempted to replicate the heat treatment used “B” which has a 50 mm gage length.

Image from [4]

Subsize specimens with a short gage length have much higher total elongation as shown below, one with 8 mm gage length and the other with 50 mm. The 8 mm gage length has about 35-38% total elongation while the 50 mm has only about 25%. This helps explain why they were unable to match the elongation of the Richtig knife, because they were using bigger tensile specimens.

Image from [5]

Low Yield Strength

The Richtig knife tested had low yield strength, even relative to its low hardness, as seen when comparing with the quenched and tempered steel. A low yield strength can mean higher total elongation and toughness. However, low yield strength also means that the steel is more prone to edge rolling when used in a knife. Which could be an issue when chopping through steel bolts. The measured yield strength of the Richtig knife was even lower than austempered 52100. Looking at the microstructure of the large Richtig knife tested, there is evidence perhaps of small amounts of ferrite or pearlite. Ferrite-pearlite would definitely reduce yield strength. Ferrite and/or pearlite is a sign that the steel was “underhardened” also called “underaustenitizing.” I wrote about this heat treating issue in this article. That in combination with the inconsistent hardness in Richtig’s knives likely points to heat treating by eye rather than with consistent temperatures in a controlled furnace. No evidence of pearlite or ferrite is seen in the small knife that was at somewhat higher hardness.

Another evidence of the large knife being underaustentiized is there is much more carbide present in the large knife which would indicate it was austenitized at a lower temperature and/or less time. I wrote an article about carbides dissolving during the high temperature austenitizing process in this article. Compare the spherical particles here:

Small knife which has relatively few spherical carbide particles

Large knife with many more carbide particles, some evidence of ferrite and/or pearlite

Therefore the difference in hardness between the two knives appears to be because of differences in austenitizing rather than the selection of austempering temperature. Why it was underaustenitzed is harder to know for sure. Possibilities include:

1. The majority of Richtig’s knives were relatively thin kitchen and butcher knives. Perhaps he didn’t compensate fully for the heavier, thicker knife.

2. Perhaps Richtig intentionally austenitized it lower for lower hardness in an attempt to make the heavier knife harder to break.

3. The use of a furnace or forge without good temperature control would likely lead to inconsistent heat treatments, both within a single knife and between knives.

I think 1 and/or 3 are the most likely, as that helps to explain the variation of hardness observed within a knife as well as the differences in hardness between knives. Heat treating without a PID-controlled furnace isn’t difficult to believe considering the time period he was making knives.

Superior properties in 52100

In the Materials Characterization article they compared to various austempered heat treatments of 52100. In a condition austempered at 300°C, both higher strength (blue, y-axis) and total elongation (green, x-axis) was achieved relative to the Richtig steel. So while the authors were complimentary of the Richtig properties, a cursory comparison with austempered 52100 revealed superior properties being possible.

Austempered (Bainitic) Steel?

Richtig developed his heat treatment in the 1920’s which is the same time period where Edgar Bain was discovering austempering and the phase that was later named after him, bainite. In the Materials Characterization article they simply chalk this up to Richtig’s genius, “it is possible that Richtig discovered the austempering process before the two well-known metallurgists were generally recognized for discovering the process” [2]. It is more likely that Richtig was using a relatively conventional quench and temper process and that the low yield strength was instead related to the steel analyzed being underaustenitized.

Dual-Phase Steel

There is a class of steel called “dual-phase” which is intentionally designed to have a combination of ferrite and martensite. The combination of those two phases gives the steel lower yield strength and higher elongation for good forming properties. One example would be Dual Phase 1180 as produced by United States Steel [6]. This steel has a yield strength of 800-970 MPa and ultimate tensile strength of 1180-1300 MPa. The Richtig knife steel tested as 830 MPa yield strength and 1220 MPa ultimate tensile strength, both values fit within the dual-phase 1180 properties. This provides more evidence that the Richtig steel behaved the way it did due to combination of ferrite and martensite rather than from austempering. While it is technically possible that Richtig austempered his knives I think that underaustenitizing is the more likely explanation of the observed properties.

Modern Knives

The Lawrence Livermore researchers said that they attempted to chop through steel with a modern kitchen knife and that it chipped. I don’t doubt that that is true. However, it is possible to cut through steel with many modern knives. For example, below I have a video of knifemaker Shawn Houston cutting through a heavy nail with a CPM 15V knife heat treated to 67 Rc. 15V is not a steel known for its high toughness, and at 67 Rc it would have even less toughness.

Edge Geometry

Edge geometry is the most important part of knife performance. Even a Richtig knife would chip or roll if taken too thin. The key to making a knife that will chop through steel and then slice paper is using proper edge geometry to survive the test. Steel can have a surprisingly wide range of hardness and toughness and still survive such a test if given proper edge geometry.

Conversations with Sid Suedmeier

Sid Suedmeier is a knifemaker known (among other things) for restoring Little Giant hammers, making damascus, and for collecting Richtig knives. He used to show his collection at the Eugene Oregon knife show. I called him recently to talk about Richtig knives. He told me that he sold his Richtig collection a few years ago, which is too bad, but Sid remains an authority on the subject. I asked him what he thought the secret was to Richtig’s cutting demonstrations, and his answer was, “Practice.” Sid believes that the heat treatment performed by Frank Richtig was not necessarily outstanding as shown by the wide range of hardness values seen in his knives (Sid has also had many Richtig knives tested for hardness). He says that his cutting demonstrations were an example of excellent showmanship to demonstrate knives doing feats you might not think they are capable of doing. Such feats were good enough to get him featured by Ripley’s Believe it or Not.

Summary

Frank Richtig’s knives excited the imagination with his fun and impressive demonstrations where he chopped through steel with his knives without damage. His legend was furthered by researchers who published information speculating that he developed austempering heat treatments before scientists, and stating that a test on a modern kitchen knife was unable to match the feats of Richtig’s knives. They downplayed the inconsistent heat treatment of his knives and the fact that they had examples of austempered steel with superior properties. An underaustenitized blade is a more likely explanation of the inconsistent properties and high elongation values of the steel they tested. Furthermore, edge geometry and a little bit of skill are the main sources for the excellent performance in chopping through steel, rather than the heat treatment. The focus on his legendary heat treatment is somewhat of a red herring. Fortunately Richtig was a stock removal maker so we don’t have to investigate any special forging processes. My Richtig knife is well made; admirable in its simplicity. I think Richtig was a very good knifemaker and his demonstrations were very effective at building enthusiasm for buying his knives. Studying his knives and marketing techniques has a lot to teach us. I will be studying my own Richtig knife further and will report on my findings in the future.

---

[1] https://clarksonhistory.wordpress.com/2013/05/06/secrets-of-the-dead-the-richtig-knife/

[2] Wadsworth, Jeffrey, and Donald R. Lesuer. “The knives of Frank J. Richtig as featured in Ripley’s Believe It or Not!®.” Materials characterization 45, no. 4-5 (2000): 315-326.

[3] Teague, J., R. LeMaster, J. Rinksc, A. Winkelmannd, and L. Bartlett. “Attempted Replication of Frank Richtig’s Forgotten Steel Heat Treatments.”

[4] Goshert, Bryan. “Effects of specimen geometry on tensile ductility, The.” PhD diss., Colorado School of Mines. Arthur Lakes Library, 2018.

[5] Thomas, Grant A. “Simulation of hot-rolled advanced high strength sheet steel production using a gleeble system.” PhD diss., Colorado School of Mines, 2009.

[6] https://www.ussteel.com/products-solutions/products/dual-phase-1180