What is Edge Stability? Part 2 – The Experiments

Thanks to Gary Creely and Sans Jeux for becoming Knife Steel Nerds Patreon supporters!

Background

Read Part 1 before this article as it covers the ideas behind the Edge Stability theory and how things like hardness of steel, carbide volume, and carbide size are thought to affect knife edges. Then you will have an understanding of what we are looking for in the experiments described below.

Test Setup

Roman developed his edge stability test described below for work done for his thesis: Schneidkantenstabilität von Messerchneiden, or Stability of Knife Edges. The test was designed to measure the resistance of an edge to deformation or chipping from an applied load. The test setup was as follows [1]:

The sample used is seen on the right and the test setup is on the left. He used a Vickers hardness test where rather than an indenter to measure hardness, a Titanium Nitride coated 2mm diameter steel rod was used instead. The small test specimen was held vertically as shown and then the indenter was pressed into the edge with a load of 1kg for 10s. Ten indentations were made at 1mm increments along the edge so that an average could be taken. The length of the half-circle was measured and then converted into a height using the following equation:

In the equation, r is the radius of the rod used to make the indentation. Then edge stability is 1/h using the average arc length, s, of the 10 measurements from each edge. An example of 10 measurements on an ATS-34 knife are shown below:

Edge Stability Tests

Roman tested a series of steels for his Masters thesis and also performed several tests on steel 1.4111 with different heat treatments. Here are the compositions of the steels tested:

For a few of the steels there were edge stability tests performed with multiple heat treatments, others were only given one heat treatment for edge stability testing. That makes it somewhat difficult to separate the effects of steel and heat treatment or hardness. Here are the heat treatments, hardness, and corresponding 1/h edge stability values (higher is better):

Effect of Hardness

Summarizing the tests from the results shown in the table below shows a clear trend is with hardness, where higher hardness meant a higher value of edge stability. This is in line with what was described in Part 1 where higher hardness means higher yield stress and therefore better resistance to edge deformation or rolling:

Effect of Microstructure

As described in Part 1, the carbide structure is thought to be important for edge stability (hard particles throughout the steel that add to wear resistance but detract from toughness). One potential outlier was one of the four ATS-34 tests that had a much higher value than the others (0.037 vs 0.022-0.027) which perhaps shows the difficulty in performing tests on a very small scale. Each test, after all, is deforming only a small portion of a thin edge. Roman speculates that perhaps the edge of the outlier ATS-34 value was in a section of the steel that avoided segregated regions full of large carbides [2]. Most of the steels were tested with only one hardness which makes it difficult to separate the effects of steel and microstructure vs hardness. There is a clear difference between the two 64.5 Rc steels, where one was a 1.1740 (0.64C low alloy steel) and the other 1.2838 (a 1.45C-2.91V alloy steel). The 1.1740 had little or no carbide present while the 1.2838 had some larger vanadium carbides, so one might speculate that the vanadium carbides led to a reduction in edge stability. However, the edge stability of 1.2562 at 63.2 Rc had a high stability value and it has a relatively large volume of tungsten carbides, as seen in the micrograph below [1]:

Micrograph of 1.2562 steel

Plotting carbide volume vs edge stability there is a slight negative trend, but the higher carbide volume steels also tended to have lower hardness. Therefore when the effect of hardness is adjusted for, carbide volume did not have a clear effect on edge stability in this dataset:

A 20° single bevel edge (which was tested) is approximately equal to a 30-40° double bevel edge in terms of cutting ability and may also be similar in terms of edge stability. Therefore it may be that the 20° single bevel edge was too obtuse to clearly reveal differences in carbide volume or size. Roman tells me that other tests were performed on specimens with a 10° edge bevel and that those showed a greater difference between low carbide and high carbide volume steels, with brittle fracture failures with the high carbide steels [2]. However, those tests were not presented in the thesis. A 10° angle with the same load applied would lead to a greater stress applied to the material, and a smaller volume of the edge, which would be more likely to show the effects of carbides.

Roman also reports that while the low carbide steels deformed plastically and had a round indentation, the high carbide steels fractured in a brittle manner indicative of chipping due to carbides. Therefore even if the quantitative measure of edge stability presented here did not indicate clear trends with carbide volume that may be because it is not a perfect representation of the response to the load applied to the edge.

Effects of Heat Treatment

Four tests were performed with ATS-34 with different heat treatments. Two of the heat treatments resulted in ~61 Rc while the other two resulted in ~57.5 Rc, and as expected the higher hardness steel showed superior behavior, even without considering the potential outlier value (0.037 seems unrealistically high). One interesting comparison, however is the two 57.5 Rc conditions, where one was given a low temper at 190°C (374°F) and the other a high temper at 530°C (986°F). The low temper condition showed slightly superior edge stability though it is within the noise of the test so I’m not sure we can conclude that a low temperature temper is superior in terms of edge stability based on that result.

Tests on Steel 1.4111

A broader set of tests and heat treatments were performed on 1.4111 which is a 1% carbon, 15% chromium stainless steel. The heat treatments performed along with the measured hardness and edge stability are shown below:

One surprising result of the series of heat treatments performed with 1.4111 was that no clear trend with hardness was observed in the edge stability tests:

Looking at the specific heat treatments it is apparent that part of the reason for this unexpected behavior is that higher hardening (austenitizing) temperatures led to reduced edge stability. Higher austenitizing temperature generally means higher hardness, so if a higher austenitizing temperature leads to a reduction in edge stability that may lead to a flat or negative trend with hardness. Here is the effect of austenitizing temperature on edge stability:

Higher austenitizing temperatures usually leads to higher hardness but lower toughness (see Austenitizing Part 2) because of grain growth and higher carbon in solution, which may indicate that the reason that higher hardness 1.4111 did not show superior edge stability was because of a reduction in toughness. Another potential effect on edge stability is retained austenite, as a drop in hardness was observed at the highest austenitizing temperature, which is generally because of excess retained austenite. So despite the use of cold treatment at -80°C, retained austenite was present in the microstructure of at least the 1100°C heat treatments, and likely a somewhat smaller fraction in the lower temperature conditions. Retained austenite lowers the yield stress even when at the same hardness [3] which could possibly lead to more deformation in the edge stability test. The yield stress is the stress at which a material begins to permanently deform, and the tensile stress is the peak stress a material reaches prior to fracture. Hardness typically correlates most strongly with tensile stress, but perhaps the more localized nature of the edge stability test leads to more sensitivity to yield stress.

The 1000°C austenitizing conditions were actually a double hardening treatment, where they were first austenitized at 1100°C and quenched followed by a second at 1000°C. Double austenitization treatments like this sometimes show superior toughness because the first high temperature austenitize can dissolve some more carbide (less carbide means better toughness) but the second austenitize gives lower carbon in solution. You can read more about these types of heat treatments in Austenitizing Part 3. However, the double austenitizing heat treatments still fit the trend of the effect of austenitizing temperature if the value of 1000°C is used in the earlier plot. In other words, the initial high temperature treatment doesn’t seem to have made a difference.

One effect of heat treatment that appears to have a little simpler behavior is different tempering temperatures. Reducing the tempering temperature increased both hardness and edge stability. Perhaps the reduction in tempering temperature was not as deleterious to toughness as the higher austenitizing temperature and therefore the higher hardness effectively increased edge stability. Alternatively, changing the tempering temperature within this range likely had no effect on retained austenite, so if it was the retained austenite that led to poor behavior with high austenitizing temperatures, that high retained austenite problem wouldn’t have been present with the change in tempering temperature. 

Conclusions and Summary

The edge stability test is interesting in that it is a test of an edge rather than a bulk toughness or strength test. Knife edges are very small relative to bulk specimens and therefore behave differently. Adding more steels and heat treatments to edge stability testing would certainly add more to our understanding. The effect of different applied loads and different edge angles are also good areas for future study, if someone chooses to take up the test method and carry it further. Roman reports that many more tests were performed that were lost in a hard drive crash, which is unfortunate. Hardness appears to be the strongest controlling factor for edge stability. There was not a clear trend with carbide volume though that may be because of the stability of 20° single bevel edges relative to double bevel edges, or perhaps because the quantitative measurement method does not capture brittle vs ductile behavior. Austenitizing at temperatures that were too high led to poor edge stability either because of reduced toughness or high content of retained austenite. Tempering at lower temperatures increased both hardness and edge stability, at least within the narrow range that was studied.

In order to ensure that this valuable research is not lost, I have uploaded copies of both the thesis and paper on 1.4111 (with permission) and you can download them below:

Thesis Part 1 – Schneidkantenstabilität von Messerchneiden

Thesis Part 2 – Schneidkantenstabilität von Messerchneiden

Schneidkantenstabilität des Werkstoffes X 110 CrMoV 15 in Abhängigkeit von der Gefügeeinstellung

---

[1] Landes, R. “Messerklingen und Stahl.” Aufl. Bad Aibling: Wieland Verlag (2006).

[2] Personal communication with Roman Landes.

[3] Rehan, Muhammad Arbab, et al. “Effect of Austenitization and Tempering on the Microstructure and Mechanical Properties of a 5 wt% Cr Cold Work Tool Steel.” steel research international 87.12 (2016): 1609-1618.