Cast Molybdenum High-Speed Steels

I. T. Tikhonov

Tungsten and molybdenum, in their effect on high-speed steel, are very close to one another; moreover, molybdenum is considerably more effective, and 1% molybdenum, in terms of producing cutting properties, is approximately equivalent to 2% tungsten. However, forged molybdenum high-speed steels, while possessing cutting properties at the level of tungsten steels, are highly prone to decarburization during heating for forging or hardening and are distinguished by a narrow range of hardening temperatures, which greatly hinders their broad use.

The manufacture of tools by casting opens up new possibilities for the use of molybdenum high-speed steels. Many types of cutting tools, when produced by casting, as experience shows, can be obtained in a finished form requiring only sharpening of the cutting edges and normal tempering. In this case such factors as decarburization and the narrow range of hardening temperatures lose their significance. For tools of this type, molybdenum can evidently be regarded as a fully adequate substitute for tungsten.

Composition and Preparation of the Melts

The composition of forged high-speed steel is determined not only by its cutting properties but also by its technological properties. In the case of cast tools, there is no need to take into account such important technological factors as forgeability and, in some cases, annealability. This makes it possible to increase the content of carbon and alloying elements in high-speed steel in order to obtain higher cutting properties [1].

On the other hand, cast tools broaden the possibilities for using low-alloy high-speed steels. As is known, many grades of low-alloy high-speed steels, under certain operating conditions, are not only not inferior to RF-1 steel but even surpass it. However, these steels do not possess sufficient universality, and as a result their mass production has proved inexpedient. When cast tools are produced by the tool shops of metalworking plants, however, in most cases there is no need for broad universality in the grades used. On the contrary, apparently even a certain specialization will be expedient—namely, the use, for one or another type of tool operating under definite conditions, of cheaper low-alloy high-speed steels that are highly productive under those conditions.

Since detailed investigations of cast molybdenum high-speed steels had not been carried out, the composition of the steels prepared was varied within fairly wide limits. As can be seen from Table 1, which gives the chemical composition and designation of the melts, the steels produced belong to tungsten-molybdenum and molybdenum steels, and, in degree of alloying—

chiefly to medium-alloyed and low-alloyed high-speed steels.

Table 1

CHEMICAL COMPOSITION OF THE MELTS

The chromium content for all the melts was taken within the limits established for the standard steels RF-1 and EI-262. The vanadium content ranges from 1.5 to 2.3% and thus does not exceed the limits accepted for steel EI-262.

The question of the influence of the carbon content in high-speed steel on its cutting properties, which is of particular importance for cast tools, has been studied comparatively little. But since the insufficiency of carbon in wrought steels for obtaining the best cutting properties may be regarded as generally recognized, in our melts we specified chiefly elevated carbon contents. In most of the steels produced, carbon is present in an amount of 1.0–1.25%.

The steels were melted in a high-frequency furnace. As raw materials for producing the melts, factory scrap of steels of grades RF-1 and U-12 and ferroalloys of chromium, molybdenum, and vanadium were used. The furnace crucible had an acid lining. During melting the metal was protected by slag formed when the broken glass charged into the crucible was melted. The metal was deoxidized before pouring by adding aluminum to the furnace crucible.

As the tool to be cast from the melts produced, end-milling-cutter blades were chosen; in centrifugal casting in metal molds these can be obtained with ready-made corrugations for fastening. The surface of the cast blades was, as a rule, clean. Although the molybdenum content in most of the melts considerably exceeded the limit of 1.6–1.9%, beyond which strong decarburization begins when steel is heated for quenching [2], no appreciable decarburized layer was observed on the blades when casting into a chill mold. It must be supposed that, also when casting in sand, the magnitude of the decarburized layer will be considerably less than the thickness of the layer that will have to be removed during sharpening.

Investigation of the Structure

Investigation of the structure of steels from the heats produced showed that the structure of cast high-speed steel may consist of four constituents: the $\delta(\alpha)$ phase, located in the central zones of dendritic formations; dark troostitic regions surrounding the $\delta(\alpha)$ phase; light fields of martensite (hardenite) and retained austenite; and the adjoining network of carbide eutectic.

The regions of the $\delta(\alpha)$ phase, being a solid solution of alloying elements in ferrite, are the $\delta$ phase fixed in the structure of cast high-speed steel as a result of the high cooling rate that occurred during casting in a metal mold.

There is no unanimous opinion in the literature on the origin of troostitic regions in the structure of cast high-speed steel. According to the investigations of Lebedev and Reviz [3], troostitic regions have somewhat lower red hardness and lower hardness in comparison with the martensite of forged high-speed steel.

The light fields, consisting of structureless martensite and retained austenite (the white phase), are obtained in the structure of cast high-speed steel as a result of quenching of highly alloyed austenite. In hardness and red hardness, as the studies of the aforementioned authors [3] showed, they considerably surpass the hardness and red hardness of the martensite of quenched forged high-speed steel.

The eutectic network, crystallizing last and consisting of carbides and austenite, may have different structures depending on the cooling rate. When high-speed steel is cast in sand, coalescence of the eutectic carbides and the formation of so-called “skeleton-like” forms are usually observed. When casting in a metal mold, owing to the higher cooling rates, coalescence of carbides is usually not observed, and the network retains its characteristic eutectic structure.

The change in the structure of cast high-speed steel as a function of its chemical composition under identical casting conditions (in a metal mold) is seen in Fig. 1. With a chemical composition close to steel RF-1, as shown by microphotograph a, representing the structure of steel from heat No. 1, all four constituents are present in the structure of the cast high-speed steel. With a decrease in the content of alloying elements in the steel or an increase in the carbon content, leading to a narrowing of the $\delta$ region, the $\delta(\alpha)$ phase disappears from the structure of the cast high-speed steel. The structure of steel from heat No. 6, having three constituents, is illustrated by microphotograph b (Fig. 1). Then, with a further increase in carbon content, the dark phase also disappears, and only two constituents remain—the white phase and the carbide eutectic. Such a structure, similar to that shown in microphotograph v (Fig. 1), was found in steels from heats Nos. 10, 12, 13, 14, 15, and 16. Microphotograph g (Fig. 1) shows the structure of the steel of high-carbon heat No. 11; this structure, just like the preceding one, consists of the white phase and carbide eutectic, but the proportion of carbide eutectic is considerably greater.

The cutting properties of a cast tool are determined to a considerable extent by the character of the cast structure. Testing of cutting properties showed that the presence in the structure of cast high-speed steel of soft regions of the $\delta(\alpha)$ phase, as, for example, in the case of heat No. 1, sharply reduces the cutting capacity of the tool. Thus, the generally accepted compositions of forged high-speed steels (heat No. 1 is close in composition to steel RF-1) may often prove completely unsuitable for the manufacture of cast cutting tools. Judging by the poor cutting properties of heat No. 6, the presence in the structure of cast high-speed ...

steel of troostitic regions, owing to their low alloying, is also undesirable. The best cutting properties were possessed by heats whose structure consisted of a white phase and a moderately developed network of carbide eutectic. With excessive development of carbide eutectic (heat No. 11), the cutting properties deteriorated as a result of the appearance of increased brittleness.

Fig. 1.
a, b, c; ×500

With respect to the grain size of the cast structure and the character of the carbide network, chill casting has substantial advantages. When cast in a chill mold, the tool receives a fine-grained structure and a uniform carbide network. The influence of the cooling rate of the casting on the structure of cast high-speed steel is illustrated by Fig. 2. In this figure two microphotographs of the structure of one and the same heat (No. 13), taken at the same magnification, are combined; here the right micropho-

…micrograph shows the structure of the knife, and the left-hand one—the structure of the sprue, which, in comparison with the knife, had a cross-section approximately half as large.

Fig. 2

Behavior During Tempering and Red Hardness

Investigation of the behavior of the melts during repeated tempering at a temperature of 560° showed that in all melts, with the exception of melts Nos. 9 and 11, maximum hardness is attained after 2–3 tempering operations. In this case melts Nos. 10, 12, 13, 14, 15, and 16 were distinguished by very high hardness, 67–68 \(R_c\), which is explained by the special nature of their structure, consisting of a white phase, carbide eutectic, and a phase harder than the martensite of wrought steel. Melts Nos. 9 and 11 increased their hardness considerably more slowly. Thus, the maximum hardness in melt No. 9 was attained only after 4–5 tempering operations, and in melt No. 11 after 8–9 tempering operations. This behavior of these melts is a consequence of their high carbon content, which led to the formation in their structure of more stable retained austenite.

In the structure of the melts after tempering, the formation of needles of martensite was observed in regions of the white phase. Moreover, the low-alloyed and medium-alloyed melts with an increased carbon content (1.0–1.2%), as the micrograph in Fig. 3 shows, acquired a structure of coarse acicular martensite. A similar structure of coarse acicular martensite is encountered quite often in some low-alloy wrought high-speed steels, for example EI-184, after quenching them from high temperatures. But in wrought high-speed steels, coarse acicular martensite appears together with an overall coarsening of the grain. Our melts, however, all had approximately the same fine-grained structure, and therefore it may be assumed that the harmful influence of coarse acicular martensite in

Fig. 3

the effect of increased brittleness of the tool was comparatively weaker.

We determined the red hardness of our melts by the method proposed by Yu. A. Geller [4]. According to this method, red hardness is characterized by the temperature of a fourfold one-hour tempering, after which the hardness of the specimens tested does not fall below 60 \(R_c\). The red-hardness values we obtained for our melts are given in Table 2. As can be seen from the table, the red hardness of most of our melts has an intermediate value between the red hardness of steel RF-1, which when hardened from 1300° ranges within 605–610°, and the red hardness of high-cobalt steel RK-10, which has a value of 650° when hardened from the same temperature.

Table 2

The structure of cast high-speed steel, when sufficiently high cooling rates are present, may be regarded as the structure of steel obtained as a result of hardening from higher temperatures than the maximum possible in the heat treatment of forged steel. And since the red hardness of high-speed steel increases with an increase in the hardening temperature, it is natural that cast steel, being hardened from a higher temperature than forged steel, also has a higher red hardness.

With regard to the influence of the chemical composition of high-speed steel on the red hardness of its cast structure (Table 2), it is possible to note a very effective action of carbon. An increase in the carbon content in a low-alloy steel (1—4—2—4) by 0.22% (melts Nos. 15 and 13) raised its red hardness from 628 to 642°, whereas the addition of 4% tungsten affected the increase in red hardness less significantly (melts Nos. 15 and 14). The same conclusion is confirmed by comparing the red hardness of melt No. 1, alloyed to the level of steel RF-1, and melts Nos. 15, 10, and 13, which are low-alloyed but have a higher carbon content. Melts Nos. 9 and 11, which showed the highest red hardness, exceeding the red hardness of cobalt steel RK-10, also had the highest carbon content in their composition. It should be noted that the carbon content in our low-alloyed

in heats Nos. 10, 13, and 15, which showed high red-hardness, is not unusual for forged low-alloy steels, for example EI-276 and EI-277. But, owing to the narrowed hardening-temperature range, the possibilities of these steels apparently cannot be fully utilized.

Testing of Cutting Properties

In choosing a procedure for testing cutting properties, we tried to create conditions that would best reveal the shortcomings of cast high-speed steel. The structure of cast high-speed steel possesses high red-hardness. Owing to the presence of a carbide network and to a harder basic metallic matrix, with respect to wear resistance it should be assumed that it will also prove more favorable for a cutting tool than the structure of forged steel. Therefore, the principal shortcoming of cast high-speed steel should naturally be considered its increased brittleness.

In order to reveal increased brittleness, tests of cutting properties in face milling of a sufficiently tough but high-strength steel ShKh-15 \((185—205 H_b)\) were adopted as the basic tests. In order not to ease the cutting conditions, the blanks were not preliminarily machined on the surface and had a thick layer of scale formed after rolling and annealing. The tests were carried out without cooling.

The testing procedure was as follows: only one cutter was inserted into the milling-cutter body, and the wear of this cutter was measured on the flank after each pass of the cutter with the aid of a microscope. As the criterion of blunting, wear equal to \(1.5\) mm was adopted. The diameter of the milling cutter along the cutting-edge circumference was 170 mm. Milling of the blanks was performed from the end face, which had dimensions \(100 \times 100\) mm, at a milling depth of 3 mm, feed per tooth of 0.292 mm, and spindle speed of 48 rpm.

Fig. 4

The wear curves for cutters from heat No. 13, shown in Fig. 4, indicate uniform wear of the cast cutters, analogous to the wear of forged ones. The test results for cutters of all heats, as well as for control forged cutters made of steel RF-1, are given in Table 3, where the heat-treatment regimen is indicated—

...treatment and hardness of the cutters tested, the number of passes until wear on the back face equal to 1.5 mm was obtained, and the average value of durability for each heat.

Table 3

RESULTS OF MILLING TESTS

on ShKh-15 steel at a cutting speed of 25.6 m/min, with a depth of cut of 3 mm and feed per tooth of 0.3 mm

The poor cutting properties of the knives of melt No. 1, as already indicated, are explained by the presence in their structure of the soft constituent—the δ(α)-phase. The poor durability of melt No. 6 is apparently connected with its low red hardness. The knives of melts whose structure consisted of a white phase and a moderately developed carbide eutectic showed good cutting properties. Thus, the knives of the molybdenum low-alloy melts Nos. 15, 10, and 13 (1—4—2—4) exceeded the control forged knives made of RF-1 steel in durability by 5–56%. Moreover, one may note a noticeable improvement in cutting properties with an increase in the carbon content of the steel. Melts Nos. 14 and 16 (6—4—2—4), differing from the preceding ones by a higher tungsten content, exceeded forged RF-1 steel in durability by approximately 40%. Apart from tungsten, in the content of the remaining elements melt No. 14 is very close to melt No. 15, and melt No. 16 to melt No. 10. Thus, increasing the tungsten content by 4% gave an increase in durability of 20–30%, i.e., it had a weaker effect than increasing the carbon content by 0.22% (melts Nos. 15 and 10). The high-carbon melts Nos. 9 and 11, which showed very high red hardness, gave unsatisfactory results in milling. This indicates that, with respect to improving cutting properties at the expense of carbon, there is a definite limit. Thus, whereas the cast knives of the preceding melts did not chip during operation and, in the character of wear, behaved similarly to forged ones, the knives of melts Nos. 9 and 11 were distinguished by very uneven wear, indicating chipping along the cutting edge and, consequently, their increased brittleness.

In addition to testing the cutting properties by milling, we also tested some melts in turning operations.

The material machined was ShKh-15 steel. The depth of cut was 2 mm, and the feed was 0.273 mm per spindle revolution. The cutters for these tests were the same cast knives of the face mill, clamped in a vertical position in a holder specially made for this purpose. Complete seating of the cutter was chosen as the criterion of blunting. The tests were carried out at different cutting speeds and, on the basis of them, a graph was constructed in logarithmic coordinates, “cutting speed—durability.” To obtain comparative data under the same conditions, a test was made of a reference forged knife made by us from RF-1 steel. This knife, just like the cast ones, was inserted into the holder and had the same dimensions as they did. In addition to the reference knife, data from tests of a solid forged cutter of section 16 × 32 mm made of EI-262 steel were also used. The graph shown in Fig. 5 shows that both control cutters gave very similar results.

The results obtained in testing the cast cutters of our melts in turning, as we expected, proved to be considerably better than in milling. As can be seen on the same graph of Fig. 5, the durability line of our melts is located considerably above the durability line of the control cutters. The durability of cutters of melts Nos. 13 and 16 in the cast state and after normal tempering proved to be approximately 5 times higher than the durability of forged RF-1 steel.

Melt No. 11, as indicated above, owing to increased brittleness did not give satisfactory results in milling; however, since this melt showed extremely high red hardness, we considered it advisable to test it in turning, where the brittleness factor should have a considerably weaker effect. The results obtained in testing cutters of this melt are shown in Fig. 6. The graph shows that, although the durability line of melt No. 11 exceeds the durability line of the control cutters, we did not obtain from it the high cutting properties expected.

did not obtain. The life of cutters from heat No. 11 is considerably inferior to that of heats Nos. 13 and 16 and is only approximately twice as high as the life of cutters made of RF-1 steel.

Fig. 5. Dependence of tool life on cutting speed.

Labels in the figure: “Heat 13,” “Heat 16,” “RF-1,” “EI262.”
Vertical axis: “Tool life, min.”
Horizontal axis: “Cutting speed \(v\), m/min.”

It is possible that the increased brittleness of this heat, with the formation during chip removal of a crater on the rake face of the cutter,

Fig. 6. Dependence of tool life on cutting speed.

Labels in the figure: “Heat 11,” “Cobalt steel.”
Vertical axis: “Tool life, min.”
Horizontal axis: “Cutting speed \(v\), m/min.”

and the consequent weakening of the cutting edge, likewise played its negative role in this case.

Heat Treatment of the Melts

This part of the work had the aim of determining the effect of annealing and hardening on the structure, red hardness, and cutting properties of our melts.

Annealing of specimens of all melts was carried out according to the same schedule and consisted of heating to 840–860°, holding at this temperature for 3 hours, cooling at a rate of 40–50° per hour to 740–760°, holding at this temperature for 5 hours, and cooling with the furnace. After annealing, the structure of all the melts consisted of a basic sorbitic background and eutectic carbides; moreover, in the structure of the eutectic carbides the annealing did not cause noticeable changes with respect to their coalescence and coagulation.

After hardening, the structure of all specimens consisted of the basic metallic mass, formed by martensite and retained austenite, and carbides. In the structure of the carbide network, depending on the hardening temperature, the following changes may be noted. Heating for hardening already to temperatures of 1200–1240° leads, in low-alloyed and medium-alloyed melts, to noticeable coalescence and coagulation of the eutectic carbides. As a result, a significant rupture of the carbide network is observed, with the formation in its nodes of skeleton-like forms and coagulated carbides arranged along its original directions. The structure of melt No. 13 after annealing and hardening from a temperature of 1240° is shown in the microphotograph in Fig. 7. With a further increase in the hardening temperature—

Fig. 7

Fig. 8

Figure labels visible in Fig. 8: “before annealing, after annealing,” “new,” “speed of cutting in m/min,” “time of durability in min.”

further coagulation of carbides and rupture of the carbide network were observed. Heating to the hardening temperature, which caused melting of the specimens, was accompanied by intensive grain growth with the formation of a secondary carbide network, which had a clearly expressed skeleton-like form.

Thus, in our case, when casting in a chill mold, subsequent annealing and hardening had an effect on the structure of cast high-speed steel that was more detrimental than beneficial. If in the initial cast structure we had a network with fine carbides of a characteristic eutectic structure, then after annealing and hardening we obtained more or less coarse carbide clusters, with an overall coarsening of the grain.

Investigations of the red hardness of annealed and hardened specimens showed that, in this respect, the indicated heat treatment deprives cast high-speed steel of its positive features to a considerable extent. In comparison with the initial cast state, after annealing and hardening the red-hardness value in all cases decreased by approximately 15–20°. Testing the cutting properties of heats after annealing and hardening likewise did not yield positive results. In a turning test, as shown by the graph in Fig. 8, in the case of heat No. 16 after annealing and hardening from 1240° we had a noticeable deterioration of the cutting properties. In milling, where the red-hardness factor is of comparatively lesser importance, as Table 3 indicates, the cutting properties of the heats after annealing and hardening remained at the same level.

Conclusions

1. In red hardness, cast high-speed steels (cast in a chill mold), owing to the peculiarities of their structure, considerably surpass forged steels. At the same time, the possibility of increasing the red hardness of cast steels by means of carbon makes it possible to obtain inexpensive, low-alloy cast high-speed steels whose cutting properties are not inferior to those of the standard forged steel RF-1.

2. There is a definite limit to improving the cutting properties of cast steels by means of carbon, since with an increase in carbon content the proportion of carbide eutectic in the structure of cast high-speed steel increases and brittleness rises.

3. Heat treatment—annealing and subsequent hardening—deprives cast high-speed steel (cast in a chill mold), to a certain extent, of its specific features and, by lowering red hardness, worsens the cutting properties.

4. Cast molybdenum high-speed steels possess high cutting properties. Therefore, in high-speed steels for cast tools that are not subjected to heat treatment other than tempering, molybdenum can be successfully used as the principal alloying element.

Literature

1. A. N. Dobrovidov, A. M. Rozenberg, and G. M. Yakovlev. Stanki i instrument, No. 5, 1947.
2. Yu. A. Geller. Stanki i instrument, No. 7–8, 1944.
3. T. A. Lebedev and I. A. Revis. Trudy TsNII NKTP, No. 2–3 (23–24), 1945.
4. Yu. A. Geller. Zavodskaya laboratoriya, No. 3, 1947.