THE UNIVERSITY OF MICHIGAN COLLEGE OF ENGINEERING Department of Materials and Metallurgical Engineering Progress Report HOT-HARDNESS OF MnSe G. S. Mann L.H. Van Vlack ORA Project 07137 under contract with: SELENIUM-TELLURIUM DEVELOPMENT ASSOCIATION, INC. GREENWICH, CONN. administered through: OFFICE OF RESEARCH ADMINISTRATION ADMINSTRATIONN ARBOR November 1971

ABS TRACT The hardness of manganese selenide (MnSe) was determined up to 9000C (16500F). Although harder than MnSe at ambient temperatures, iron is softer than MnSe at high temperatures. This accounts for the globular shape of MnSe inclusions in free-machining steels. This is in contrast with MnS inclusions which deform more than the surrounding metal matrix. It represents a favorable feature for the use of selenium in free-machining steels.

HOT HARDNESS OF MANGANESE SELENIDE (MnSe) G.S. Mann and L.H. Van Vlack Previous work by Chao, et al. (1) provided a comparison between the hot-hardness values of MnS inclusions and unalloyed ferrite in steel. Those data covered the temperature range of 200 to 9600C (Figure la). The data were interesting because the relative hardnesses of the two phases have a significant effect upon inclusion deformation and fracture. (2,3,4) Subsequent work by Riewald(5) showed that the hardness of MnSe was less temperature sensitive than MnS in the -70~ to 1350C range. Although softer than MnS at ambient temperatures (Figure lb), an extrapolation of Riewald's data indicated to Aborn(6) that MnSe would be harder than iMnS at steel-rolling temperatures. He proposed this as an explanation for the more globular shape of selenide inclusions than of sulfide inclusions in commercial steels. While plausible, the authors of this note felt that the extrapolation values should be replaced with experimental data. Hardness data were rerun for ferrite (Ferrovac-E) and manganous sulfide (MnS) from 200 to 800~C. Data were also obtained for manganous selenide (MnSe) from 200 to 9000C. The present hot hardness testing apparatus utilized the same furnace as Chao(1) used. It was evacuated and backfilled with purified argon to restrict sample oxidation. Titanium "getters" were used in the furnace to scavenge any oxygen leakage. Improvements were made in the load application. Specifically the Vickers

machine was replaced by a dead-weight load counter balance over ball-bearing pulleys. Depending on the temperature, loads were chosen between 1200 gms and 200 gms with sensitivities to less than 1%. The hardness was calculated(7) according to DPH = 1.85 L DP H = d2 where L is the load in kilograms, and d is the average length of the diagonal of the impression in millimeters. The results are shown in Figure 2. In general the data corroborated the previous work by Chao, et al. (1) There is a slight modification of the high temperature hardness of iron. The current data are preferred in view of the insensitivity of the Vickers machine used by Chao to the light loads required for the higher temperatures. While there is some evidence that larger loads reduce the calculated hardness values a few percent, other unidentified variables are equally if not more important. Figure 3 shows the lines of central tendency for the hardnesses of ferrite, MnS and MnSe from the present data, plus the a+y hardness shift detected by Chao. The crossover of the MnS and MnSe curves predicted by Aborn (6) was observed. Thus his hypothesis still remains plausible, - specifically that the globular selenide and the elongated sulfide inclusions shapes are a consequence of the relative hardnesses of these two phases compared to the hardness of the iron phase. Caution should be used, however, in applying this conclusion to most steels since selenium-bearing steels usually contain high chromium contents. Kiessling(8) has shown that chromium affects the hardness of the

selenide phase, and of course, chromium will harden the metal phases by solid solution. REFERENCES 1. Chao, H.C., L. H. Van Vlack, F. Oberin, and L. Thomassen, "Hardness of Inclusion Sulfides," ASM Trans Quart, 57 (1964) 885. 2. Chao, H.C. and L. H. Van Vlack, "Deformation of Oriented MnS Inclusions in Low-Carbon Steel," Trans AIME 233 (1965) 1227. 3. Chao, H.C. and L. H. Van Vlack, "Fracturing of Oriented MnS Inclusions in Steel," ASM Trans Quart, 58 (1965) 335. 4. Chao, H.C. and L. H. Van Vlack, "Flow of Steel Near Inclusions During Deformation," Matls Res & Stan. 5 (1965) 611. 5. Riewald, Paul G. and Lawrence H. Van Vlack, "Slip Behavior and Hardness Indentations in MnSe and MnSe-MnS Solid Solutions," J. Am. Cer. Soc. 52 (1969) 370. 6. Aborn, R.H., "Plastic Deformation and Fracture Characteristics of Compounds of Manganese with Group VIB Elements," Selenium and Tellurium in Iron and Steel, Proceedings 16 April 1969. Stockholm. 7. ASM Handbook, Metals Park: (1948), p. 95. 8. Kiessling, R. and N. Lange, Nonmetallic Inclusions in Steel (Part II), Iron and Steel Institute (London), Publication 100 (1966) p. 147.

200 I00 80 60 ron 50 40 N 20nSe -_ n (I) (I) ~ ~ ~ ~ ~ ~~I w 2 Cr0 mJ 8 - \\ 55~~~\ \ 4 3 F. I a. Hot-hardness curves by Chao( for MnS andO-Fe. 2..... I...l...I.... I 1...I... 0 200 400 600 800 1000 1200 TEMPERATURE, ~C

E 80 2 MnS -f 5o,, 40 z 30_ 10 8 _ roF. lb. Hardness va ues for MnSe near room t empprature' comnared to WMng. -200 0 200 400 600 800 1000 TEMPERATURE, ~C

(G 60 nMnSe 12:: 0 3: 30 1) C 20 o 0 1 0o0 0 LOAD - 80 t ho 400 GMS. 60 x 600 LLJ 50 o 800 z x 1000 " 40 01200 Io 30 8 MnS E 20 E 100 80 U, 60,, 50 z 40- \Fe at w 20 1O - Fiq. 2. Results of this study. 0 8 0 200 400 600 800 1000 1200 TEMPERATURE, ~C

200 100 80O Iron 60 50 40 20 6er MnSen cr 0 1 - \,,!',< \ Fig. 3. Comparative high-temperature \ 6 hardnesses of iron, MnSe, and MnS based on data from th s study p I us the \ 5 — a-Mahardness differential of the pre4 vious study. These relative hardnesses could account for the different shapes of MnS and MnSe Inclusions in 3 hot-rQ led steel as discussed by Aborn 6). 0 200 400 600 800 1000 1200 TEMPERATURE, ~C