M—a‘r. w -0 - fi.‘ 7 .‘fl'..fl~.«-“0 -..—.?." 0-.» > ’ I _ C 3:11.: . h , “4% 3;. 33.33.3733 .I , [(’~'~- ~ “'3 éfiifit'mf’A‘Vfii '3" “.0 3 0 '.~. 5“. .. ..4~' u ‘ uppqn'n " n p..;gu Dr'- Vr’“ :' " '4‘ g k: 't. H “311 It .& v -;’._-35..-3\.-. ’E {.2 .vut‘flf “"‘ " _. . ‘0‘ , -. o. I w L‘ h . o 3‘.” ' ' . 7 I f V 3 1 9' .1. 3;. .4 Jr I I. ’1" '1': _ . . ‘ .. 'I ‘ "é‘n -. {.du Lfk§l7&‘ .-).. ‘ '1‘. -, t r I ‘45” ; ‘- ' ~ . ' 1 ‘. I‘ I; W 4* _ ,- ‘3’” 32-.-’3’ , [8313“. . ' ’1; - @?L‘E E!" 1\:?:}"}’ ”r. .‘f’ ‘c v . "J"€IJ:”“ ..: . .. I‘ I. '. " 3 2 '.' ’1 :.-‘ 21W- ~- ’ 5,..321533w13311,3333 .I j r. ' .' 3 lg; . . ‘4 &\. ,-'f=‘¢‘ 7 ’4‘“- fr: ‘. u ’2 3‘ I 3’ 1 - v0.0.“ 0' : . .‘Jr a. Q . I 9".' .; 310*. " Q, .~'o . I l t I‘ ll" ‘ ”c“ H w; l \ ‘V :’,’3.;; :. ' ‘1. i. Y ... Lifi‘fl ‘gl "\Of:’lv.’;’l ‘!‘| 33'!‘ I y’t‘r Lu‘ ‘~)~ . . r. 3.‘ A 3...»..43‘! '3‘. .; .. . : -‘«..__._ o .. ‘ ~... .. ,, . ~- .-,fi‘§:{fi_{$,;;£ ”'3‘; ‘f; ’P’F’figfi 3322. €711.“ _| 3.7, 1L." (1 . _-..'3‘.';‘1:.11.im§.}',;k {fin}? 3.)! :2 a“? I “."3' 3 3-?“ | 7‘ v t I". " ‘I' ‘4 .I vl.‘;l’- .13. I" .0. 3-. ' "i‘ I '~-. ""- ‘- ‘."“" .' ’ .‘I‘3 ‘.>‘ ".a‘kav!" " "I." A" .“ J..4"<5‘:‘Il‘ ~ ~‘»n '1‘ 7".”L' ‘ ".' ' ""3". 19"“: SN; .0 It. \ ' H .'~ \.v .__ ' .' ~... . 5.: 7-3 3. ,uo . . , . , .--,. . L, - “5’"'{£Jo’£l' 3'.“ :yggb’ ,I 1; 2'1“. .‘ 0-" o :43" "1’12"?“ ...~‘:‘.:IJ"_~I%€F'?;‘:. u’,’ "'h“ . r3, a-‘“ ‘H 3.“: ' 1‘. ' ‘-‘\' "’ ' f 'M .- ”3. ‘Qt ‘ -,. "'.‘\ 3' .f‘”‘ ' r‘. '3’~ )2 (‘ .-_‘ .’;'Ll~;,:i" ‘,_"'f .‘x "Hf; a“, ‘Vfifi' 4 ‘kjtf' hi: *‘L‘F‘ ‘ .1 § 5 ,, 91;; ' ~’. " |‘.’ ’k 3"}. .‘J‘l ' A “Jul; "1"» " 4' ' m 3, m-"az. aw .; . n ’ “£~‘“" ~ ' 7’. '.'.>:-’ I h." ~.‘;I‘.' u. 5“ O, “a?“ ’ z’vld"u5¢'~"' “I’ "‘3‘ “.3" “-"I‘II‘" "G‘A h'l'.‘ J “ \ .'3 . -."h '.«t '-«' . ‘. . .4 1 “I“. I. "53- ; "(‘G_...u~,t . “‘3 ‘.“L 3 ‘ ‘P-\ ‘ ‘a .‘.t. *-'w’. .. -".".~.I'-'s H-Pfi “41;. L‘? ‘Ahwkii‘sf' u . 3.?“ .‘-.}-.tul 1.‘ ~__. }. O l 0,, \ \ A .. -.._' ."3h'fut'whi1“"“13'13‘ VAN-.3 5!. ’xil hm 1w“; L . "3.34:... ., "'. ‘. '~'.'.' a . . "6. ~37 '31' 93' ' ‘c 5" silt. “a,“ 'Q'S'N‘mr ‘1’?) ‘7“ ‘ L ' x; 3”,...” '3. _‘3 ‘y I.N.'X-l“gf'w. , I” 5,.- .Izr, .I\.. ‘3‘ “.15.. | . #:K'TJ‘H.I‘;11".”:‘s'fiw$15!? ;? {.1 "$2,716,333 J . 4m: . '4 ~~""b“}$§ . "‘ 3’s ‘3 "'5" "4?» 3 ' ' “I K’ )1"‘~'.' K '1‘“ r -' ‘ 't 5. ~r‘“_,3 ‘0.” ‘3. A. """"N” I ,_ r u . .h .kfi fl. _ Kg"? '1?!” ”(Eu ("3);WJLQIL 3': J \,1:§\ ’37., {Kits .1: 0L“, "’§i"?: I; :- $1 “‘ ’wxlef‘ls; {I“\l..j.;:'.’_‘zr. L‘IBVIQ’ ‘.L‘ P',‘A“F~.\"1‘4' d~'§'..h‘{ "Qh‘éf. ’3 o“ .: I’m .'- 3 'x'.... \u 'g‘ I... . u. 3 "-54} f“. 1-. 5.1.1 , 3:“ :0 [ti-'0 . 7». . . ”3?. ,x ' If“! :‘\.:‘, N“ A). I“ "no I "T ‘ l .c i . 3 ”«'.;~’:-7:2&’: *3 7.4 t ‘. .-J g! "' ‘M 3‘ O 332.“, A.’ 0.?) N . 1’ f. ’I. ’ D 3“ f ‘ 1' 'fi . , . ' P" 0‘..‘v é ‘.' ’ -. . HI.” ' I. ' "d ‘u M v‘ ”$3.1. .1 w 5. ' . . ‘ . s.’ ¢ I 1" , - p a “1 ~.4 '\g"'t K I ..' v v -, , . L a. ‘ 31' 1.33”?“ S ~ ‘0 ,. ‘- ‘ .3 3?“ 3 wt .’ ~ .35" ‘ o. ' 0 I N! ’, ' 3 4 . its“ 0 V.‘ “q, |. “._.“j A’. ~) I ul’k?“‘ v. '.‘I I-‘-. t ‘ is. .,,. 5% .‘fi n Io.‘ . ‘, ”"ft’u' 3.4." l 'v 23" ‘34 “n 1 ' "4 V ""81; -\-s' s r - ‘ "3-7 3 1,.» v v.15; .. N“ u- ‘w' '- s" '1’" .3 ire? ”9""‘13- — '3'. ‘ -.- ' ' I '~ . K ' ‘c' - . _ “I . _ '.- 34.. 2.. .‘.‘.", Jl': ( "1‘ u¢g '0‘ ".?i:4.3'1,':3‘0~1"“..‘l'l'.\.."\'. »‘ . ..:;I‘“e. t;!§f,...,' }."’"-‘§ l '(b‘k ‘ "QM... ~ 00.. nig\.‘*‘; ‘;;.‘;$ 410:, rs... ‘1')1";\u: “.0, :l'“.“"~'.;ofir" ”‘5' 9: 3-: I - h ‘3' ' N " " I u 3. ‘9.» ’ .' K ‘3‘ .3ng! 'I‘2 ’-‘-y"“){wi‘£; ‘QZPSFNS‘5 “ "-1. ’3‘" 4h. ' ffiggififi?"~’g"§g‘§:#:ltle-I." .r'finl ’ “ ‘ ‘ .2 'W-‘W ‘ ' .*"v.a:°~ . 2v. 0 ‘ ..~ ‘. . to “ l‘ ‘h1 NRA.“ -.: '4.» ‘ '~.-“ gs 3:1?“ . 3‘1,“ ‘-'QT\-‘ I,“"?Ay. ’.9.\.'u‘¢ § _ Q, ‘V. 3‘. " .sof‘ . ' IH _'u9 ‘::”__,.. 4" . v trouble heretofore. It has been stated that the Uviarc and 5400 3 filter could not be used for focusing whenever an ocular was used in the microscope. The image produced under these conditions could not be seen clear enough to focus. Consequently it was necessary to use the tungsten filament to increase the intensity of the light on the ground glass. The conclusions drawn from the above are that an image focused in light of 5400 2.13 in focus for ultra violet light of 5650 fl.regard1ess of the source of light; an image focused in yellowbgreen light is not in focus for ultra violet light of 3650 X. This difference in focal point for light of various wave lengths does not seem.to be as pronounced for the oil emersion objective as for the 6 mm, A series similar to that shown in Figure 10 was taken from.a non-ferrous alloy of capper and aluminium» Nothing was gained by using ultra violet light (See Figure 11). ' Figure 12 shows two photomicrographs of capper. Photo 12a was taken with ultra violet light and 12b with the 1.7 U. V. objective and yellow-green light. Here again the short wave light has produced a.more nearly flat field and finer separation of grains. Thus far the author has used only Metallographic plates. ESeveral attempts were made to focus and photograph using Inercury green filter. Some of this difficulty was eliminated by using Wratten M plates which are more sensitive to this QYpe of light. This can be seen by comparing photos 'a‘ andi'b’ Figure 13. Photo 15a shows that the‘Metallographic 11. ') ".‘V'fl ) 1 .. Q J 0 -.,~ ‘Arfi . II a '7 . V '. “.Y‘ P‘ r. 4 r_ '_ a a w “.1 C.) : '. .L 1 L. .- \_.. « ‘4 . . 7017+ 7-1“ 4-... ...; ‘ .‘ . V111“ ....-. f__ in . - _ O L' a r— r* ‘ . . I _' .. Ah \‘ f'. ‘ . \- I \ . q .I.IJ.| fur. : 1 .1? ‘r k d o ...? o « . ’j/‘c‘ 1h“ P F‘ '1’. ‘vcl-.1. ’00 .v:0 “C 7,,- T1- u‘r fllfe. LchO . . fiIFOSTTS ”5 mi“ ‘- - | r' “ .4. “(,9 C‘ (Ix. ‘0‘; can-1M» '1‘} ’V- _ Tr U ' -- . 90 t. ' '. um 6 a. a Wits ‘3 f +‘ ’ - q'..‘. 1"“ -(_l r- -c_.,v‘.{"’:'1 *111-1‘4. 4—13 .‘ _ F3 ." glu'ln‘s 1176 {r .' :12". .. a... .-. 9.. era—“JV . ... v \t 7... 1:. ‘3 plates are useless in light of this wave length. Comparisons were made between the two plates using ultra violet light. Representative photomicrographs are shown in Figure 14. Out- side of the fact that the Wratten m plates'are faster. very little difference can be seen between the results obtained with the two types of plates. The Metallographic plates are not sensitive to red light which makes them much easier to handle and develop. This is especially desirable where long exposures are required as in ultra violet photomicrography. l2. 1.P+r‘ 1 -p"y~ " n .-h‘. ..... 09¢! '1, w w J- .L it"uf) 0 Figure 14. Relative Sensitivity of Photographic Amultions to Liggt of 3059 R “'r v - T '»».*.~+h .u;~vb’ “CMHU 1' w... x .. fut“ \...\.1.. ... til 3.. .\ .\ ... .. .x . . .. . x ..on 7‘ '8 I .\ as Kw»)? S L. .t . 351‘“ a- xxx . N. X t . V’ n“”u~lfl.\.§\\ .AMNNX*& xw‘sfl ~. . . ..m....\._..,.\.....\.m..\.. ...n . ..\.\.\.......\e\¢~.\vt. . . . \..... ..\..._...... ....t . .ls. . ;.Wnt.f.m.ts1 a. . . ... DISCUSSION Photomicrography with ultra violet light requires a very delicate touch. In the first place. the specimens must be polished much flatter than is necessary for ordinary photomicrography with yellow-green light. Etching is another problem.to be borne in mind. Since ultra violet light has a high disolving power it becomes necessary to use an ex- ‘ tremely lightoetOh to bring out fine lines present in the structure. Quite some time was spent in developing this etch. NOthing can be said about the precautions necessary other than stating that weak etching solutions and clean- liness are important factors. The:mercury green filter (5400 R) supplied with the equipment for focusing is very dense and admits very little light especially with an oil emersion objective. It was found impossible to use this filter for focusing with an oil emersion objective. It was extremely difficult to use even with the 6 mm. objective. The field was too dark and hazy to be brought to a sharp focus with any accuracy. This was especially true with anything but a 7.5x Hyperplane ocular. As a result, practically all the focusing had to be done with the standard Madza Lamp. Both the 5400 R and yellowbgreen filter were used in focusing (See Pages 10 and 11). This necessitated changing the lamp before and after each short wave exposure. Besides the inconvenience -- ‘13.. encountered in so doing, it was very difficult not to jar the microscOpe and thus throw the specimen out of focus. EXperiments with different makes of eyepieces showed that Homal and Compensating eyepieces have a high absorbing power for ultra violet light and are useless in this work. Hyperplane oculars did not absorb nearly as much and were found to be more satisfactory. The 12.5x ocular absorbs nuch.more than the 7.5x. Even with these eyepieces a very short bellows was required to keep the time of exposure within reasonable limits. I Ultra violet light increases the time of eXposure enormously. The ocular seems to lower the efficiency still more. This can be shown by referring to some of the times of exposure listed on pages 8 and 9. When no ocular is used the time of eXposure is increased about two and one-half thmes with ultra violet light over that with yellow-green light. When a Hyperplane 7.5x ocular is used the ratio of exposure is increased 30 to 40 times depending on the type of surface and the amount of ultra violet light absorbed by it. Again a 12.5x Hyperplane ocular increases this exposure from a few minutes with yellow-green light to 6 to 7 hours with ultra violet light. Apparently a considerable amount of this increase in exposure is due to high absorption of the short waves in the ocular used. These comparisons are only relative since some specimens absorb much more light than other. and thereby cause variations in exposure. Some exposures of 12 to 15 hours produced plates to light to print. 14. CONCLUSION Very little time has been spent in investigating various types of structures in this report. The author has spent most of his time with specimens of annealed steel; the primary object being to find if photomicrography with ultra violet is practical from the standpoint of increased resolu- tion, economy, and technique required. It is possible to produce finer lines on a photomicro- graph by using short wave light. Photomicrographs included with this report have shown this to be true. Possibly it could be shown more convincingly by using some quenched structures instead of annealed steel. All this being true the author still believes the equipment manufactured by Bausch and Lomb Optical Company to be extremely inefficient. Why did the manufacturers emphasize the use of quartz frmm the light source to the object and pay no attention to the Optical system from the object to the photographic plate? It seems they forgot that the light was reflected from the specimen back through a plane glass reflector, then to a stellite prism and an ordinary eyepiece. The author doesn't know how much ultra violet light is absorbed in passing through this reflector and reflecting prism; howeveru the extreme increase in exposure caused by using certain oculars seems to be proof enough that they absorb a considerable amount of this short wave light. It is reccommended that 15. the apparatus could be greatly improved by making the entire optical system of quartz or a very high grade of Optical glass which has high transmitting power for ultra violet light of 3650 a wave length. It is also reccommended that some change be made either in the mercury green filter or the Uviarc or both so they can be used for focusing. Although the apparatus as used gives slightly better resolution than standard equipment the author does not reccommend its use for routine work. The technique is too involved and the exposures too long. It has been impossible to get any photomicrograph at 2000 diameters with short wave light. The plates were slightly out of focus and much too light even after 15 to 18 hours exposure. This is much too long for the added resolution that can be obtained. 16. REFERENCES Optical Instrwments for Examining and Analysing Ketals. Bausch and Lamb Optical Company. Ultra Violet Photomicrography Accessories- Bausch and Lomb Optical Company. Ultra‘Violet Photomicrography. A. P. H. Trivellt. Ber. VIIIth Intern. Congress of Photography, Dresden, 1931, 338-40. 17. PART II . THE STRUCTURE OF STEEL AT ELEVATED TEMPERATURES UNDER A REDUCED PRESSURE INTRODUCTION Many attempts have been made to study the change in microstructure of steel when heated to temperatures within the critical range. In.mest all cases the steel has been heated either in a neutral atmosphere (usually hydrogen or nitrogen) or in a vacuum. Some workers have heated pure iron and steel in a vacuum then admitted a small amount of etching gas such as chlorine. After a few seconds this gas was pumped out and the sample cooled in a vacuum. In all ' cases the structures present after cooling were assumed to be those present at the highest temperature reached. It was also assumed that the presence of a neutral gas did not alter the surface changes occurring within the transformation range. The results of these investigations can be summed up in a few statements. The surface of a piece of polished steel shows evidence of buckling after cooling from within the transformation range. It is believed to be the result of a volume change occurring at the critical temperature. A surface configuration or pattern was present to which various names “heat etching“. 'heat relief“, and 'vacuum etching“ have been applied. Specimens cooled from above the A3 critical showed two networks, one superimposed upon the other. Rawdon describes these structures as being those present at the A3 and A1 criticals respectively. He con— cluded that the straight-sided polyhedral twin crystals were 1. a record of the gamma-iron present above the A3 transforma- tion and that the second network was that which prevailed below the A5 transformation. The volatilization of surface metal has been observed by several investigators. This volatilization causes decar- burization of a very thin layer of metal at the surface. The phenomenon is more pronounced at the grain boundaries than over the face of the grain. This has been shown by the persistance of slip bands upon heating, also etched samples show a slight roughening of the ferrite layer after heating. Other investigators have found indications of volatilization by pressure measurements. It has been reported that vacuums of less than 35 microns could not be obtained in furnaces containing steel above 10000 Fahr. while at room.temperature vacuums as slow as 15 to 20 microns were consistently main- tained. This increase in pressure occurred within a definite temperature range. Rawdon advanced the idea that volatilization was more pronounced at temperatures within the A1 transformation than at higher temperatures. He explained this on the basis that the diffusion of cementite is increased with temperature. Although the rate of volatilization is perhaps increased at the higher temperatures its effect is not as noticeable due to the increased rate at which the cementite is approaching the surface. Sauerwald, Schultze. and Jackwirth followed Rosenhain's procedure and interpreted the evidence of recrystallization 2. as that of the filmed surface layer which was produced in polishing. Howe gave another eXplanation based on the theory that steel cooled from a high temperature expells the free con- stituent (ferrite or cementite as the case may be) to the outside of the mother austenite crystal which forms the ex- terior. By repeated heatings the layer is augmented to such an extent that the specimen has the appearance of being de- carburised. Rogers and Fort have reported one of the latest pieces of literature on this subject. They heated Armco Iron in an atmosphere of hydrogen and recorded the various structures that occurred during heating on a motion picture film. They showed the A3 transformation as a wave passing over the sur- face of the specimen. They concluded that it was the result of a volume change at the surface. The present work has been performed in a vacuum.of less than one millimeter of mercury. The structural changes occur- ring at the surface of a steel specimen have been studied as they appear when the metal passes through the transforma- tion range. Photomicrographs have been taken to record many of the structures observed. The following pages describe the technique and procedure which was employed. APPARATUS The experiments were carried out in a small electric vacuum furnace. A line drawing of the furnace is shown in Figure 1. It was so designed that it could be placed on the stage of a Bausch and Lomb microscope. The holes drilled in the base provide entrance for thermocouple and heating element leads. A fourth hole drilled horizontally through the base was connected to a water line for cooling purposes. The specimen set on a cylinder of quartz supported by the base of the furnace. Temperature measurements were made with an Iron Constantan thermocouple made of B. at 8. Gauge 25 wire which passed through the quartz cylinder and into a hole in the steel sample. The end of the couple was about one-eighth inch below the polished surface. The temperature of the polished surface was assumed to be that recorded by the couple. The top of the furnace was equipped with a glass window which made the specimen visible from the outside. The height of the furnace was such that the distance from the top of the specimen to the outside of the glass was less than the working distance of a 16- mm. objective. A gravity Beal was maintained between the tap and base of the furnace by grinding the contact surfaces. It was found necessary to pass a stream of air over the glass window to keep it cool while observations were made. The electric heating element made of Chromel resistance Wire (B. 8c 8. 18) was wound in a Spiral. This spiral was 4. Figure 1 . Line Drawing of Vacuum Furnace. LI ‘\ .- a mam\ ....... fr. ...... ........ 7 V § _ m "m V / m , /a MW/fl W V . m . ,/ /7//i about one-fourth inch larger in diameter than the shank of the specimen. It was placed around the shank and insulated from it with a good grade of mica. The element was enclosed in a silica tube and packed with mica and asbestos cloth. The entire element and insulation had to be heated in a vacuum several times to drive_off all gases before a specimen was placed in the furnace. Figure 2 shows the furnace resting on the stage of the inicroscOpe ready for use. It also shows the following pieces cxf apparatus that were used in the investigations: Cenco Hyvac Pump Mercury'Manometer Leeds and Northrup Potentiometer Carbon Arc Variable Resistance Electric Heating Element Steel Specimen Air Line Water Connections An A. C. 110 volt 60 cycle current was used for heating the specimen. Figure 3 is a closeup of the furnace and llixcroscope ready for use. Figure 4 is a picture of the fur- nace with the tap removed. showing specimen and heating element in position. ' Specimens of low. medium, and high carbon steel were ttitlcen for examination. All samples were thoroughly annealed 5. Figure 2. Figure 3. Picture of Apparatus Picture of Vacuum.Furnace used for the Investigation. and Microscope ready for use. Figure 4. Picture of vacuum Furnace ‘with TOp removed. all} 11 I». a. ...; O? ..a ... en 7. .. f. - v‘a : . . r v.2 at 18500 Fahr. to produce a coarse grained pearlitic structure. They were polished in the usual manner with Alumdum Grain Size 240, and 520, and Levigated Alumina. The polished SpeCif mens were then placed in the furnace and surrounded by the heating element. It was found that a small disc of copper, placed over the specimen and element, kept a considerable amount of heat away from the glass window. It also acted as a collector for various gases given off while the steel speci- men was being heated. After the furnace was finally assembled the vacuum pump was started and allowed to run some 20 minutes before the heat was turned on. During this time water, air, and light adjustments were made. Various heating rates were employed. The minimum.time necessary for the specimen to reach a temperature of 10000 Fahr.was about 5 minutes. Above 10000 Fahr.the rate of heating decreased very noticeably due to the small temperature differential between the specimen and the element. Many times it took 30 minutes to reach a temperature of 1600° Fahr. 6. EXPERIMENTAL The original structure of the medium carbon steel is shown in Figure 5 and again in Figure 6 at 2000 diameters. It consists of large grains of pearlite surrounded by excess ferrite. The pearlite is in a lamellar condition character- istic of annealed steel. The sample was repolished and heated to 1550° Fahr. in the furnace. The surface was unetched. After cooling it was composed of two distinct networks one superimposed upon the other (See Figure 7). The grain for- mations were perfectly independent of each other and had different points of focus due to the difference in the height of the boundaries. The boundaries between the grains in the uppermost structure were much wider than those of the less distinct network. Figure 8 taken at a lower magnification shows that the entire surface has been roughened by this volume change which occurred at the elevated temperature. Figures 9 and 10 are photomicrographs taken from.a sample that had been etched before heating. The same general structure was present except that some portions of the sample had a.matted appearance - probably a.roughened condition produced by the etchant. At 200 diameters the surface con- tained many light areas as shown in the latter photomicro- graph. lt can be seen that the original structure has been entirely obliterated. Pearlitic areas were missing and the entire surface was brilliant. The dark areas correspond to those occupied by the pearlite grains and are roughened more 7. Figure 5. Original structure of e.hypo-eutectoid steel. Figure 6. Same as above at 2000 mag. . ‘ I".' t C. ' . J.\".'U".““ , '9' A .~ “as ’15 .A‘fig.‘~ .. may . Figure 7. Unetched sample of hypo-eutectoid steel after cooling from 1500°r. 1800 nag. ’1.er 8 0 Same as above at 200 mag. v _I ';..u .‘\' - I‘\ 3!: Fit'dw.‘ " I 1.5 . 38' 'v ”-fi/ - v f’Jf- '. F‘J‘qx 9 MM Figure 9. Etched specimen of hypo-eutectoid steel after heating to 1.500%. and cooling. 1800 mag. Figure 10. Same as above at 200 nag. . .‘v‘ ‘ r‘ M} a}. than the light areas due to the etchant and removal of the carbon from the surface. Figure 11 shows the same surface after being lightly etched in nital. The areas originally occupied by the pearl- ite have nearly lost their identity. Straight line formations and dark globules on the surface mark the existence of minute particles of cementite which were originally covered with a very thin film of ferrite. Notice how the boundaries brought out by heat etching still persist throughout the etched areas. The spotty condition of the pearlite is shown more clearly ' in Figure 13. This photomicrograph was taken by focusing on the pearlite thereby causing the ferrite to appear rather smooth and even. Actually the same roughened condition ex- isted in the ferrite as was present before etching. Generally a very light polish removed this ferrite layer and restored the normal structure of the pearlite. Figure 13 shows the same surface after a very light polish with lev- igated alumina. The double network was entirely obliterated and the surface was composed of normal pearlite and ferrite.' Figure 14 is the structure after a longer polish on the same wheel. Both.surfaces were etched. It is evident that heating in a vacuum.removes carbon from the immediate surface of a polished piece of steel. Let us go back and study the above stated reactions step by step as they occur when a specimen is heated in a vacuum. A sample of coarse grained steel (See Figures 5 and 6) was polished and heated to 1250° Fahr. Figure 15 is a t 8. Figure ll. Surface shown in Figure 9 after etching in nital. 1800 mag. Figure 12. Same as above at 200 mag. Focused on dark areas. a...vir-...r .- .1 - L I Figure 13. Heat etched surface (isoo°r.) of a hypo-eutectoid steel after receiving a light polish with Levigated Alumina. Etched in nital. 200 mag. Figure 14. Same surface as above after polishing wdth Levigated Alumina for 2 minutes. 200 mag. .J l (A . U... h.. . . . ..P i. .L !\ a .. .. u o ?"\) r; " e‘. ' _; ‘7' photomicrograph taken while the Specimen was at this temp- erature. The surface was very brilliant and free from any visible film. There were two points of focus: one looked as if the ferrite around the pearlite had expanded and raised above the surface; the second was a result of a change which started at about 11000 Fahr. This structure is shown in the photomicrograph and consists of entirely new grains character- istic of the recrystallization which occurred. It was ex- tremely difficult to center the light and obtain clear neg- atives at.e1evated temperatures. The focus continually changed due to expansion of the sample. Also the red light from the specimen tended to fog the photographic plates. The surface remained brilliant after cooling and gave 23 very clear grain formation as shown in Figure 16. At 100 diameters it looked similar to an etched surface of pure ferrite. A.magnification of 800 diameters revealed a pecul- iar formation in the grain boundaries (See Figure 1?). They no longer looked like etched boundaries. A dark constituent was present which at a.higher magnification looked like a mass of small spheres. Apparently a reaction had occurred at.the boundaries more rapidly than in the interior of the grain. ' Figure 18 shows the same surface after etching 10 seconds in a.l% solution of nital. lMany of the original Peaaflitic areas are darkened. At a higher magnification (Figure 19) these areas showed no signs of the original laminated condition of the pearlite. The structure was 9. Figure 15. Figure 16. Photomicrograph of a hypo- Same as above after eutectoid steel taken at cooling in a vacuum. 1250°F. The specimen 100 mag. was not.etched. 100 mag. Figure 17. Same as above at 800 diameters. . .... .Hs. fix“ ~_ 3 ..-. .rL Wu. .1; C 7.0 (.- _. .Iem _ c‘". ‘1. . n~ - In. very similar to that found in the grain boundaries before etching. The ferrite has a velvety appearance which was characteristic of the surface when viewed in the microscOpe. The network in the ferrite is the beginning of a second volume change previously shown in Figure 6. The specimen was ground off and repolished. Figure 20 shows the structure of the interior. The large pearlitic grains were no longer present showing that the entire sample had gone through a recrystallization. Figure 21 is another sample of medium carbon steel etched after heating within the transformation range. Notice that the pearlite has been coalesced and the greater portion re- moved from the surface. The concentration of the cementite is much higher in the grain boundaries than over the interior of the grains. A sample of steel having a structure consisting of pearlite and ferrite as shown in Figure 20 was polished and rapidly heated to 1170° Fahr. in a vacuum, Figures 22 and 25 are photomicrographs at 100 and 800 diameters respectively and show the surface after cooling to room temperature and etching in nital. Almost all of the pearlite has disappeared leaving brilliant grains of ferrite on the surface which are much finer than the original grain structure. Practically all the carbon that can be seen is concentrated at the wide boundaries between the ferrite grains. This indicates that carbon leaves the polished surface simultaneously with the first signs of grain formations or possibly before any visible 10. Figure 18. Figure 19. Same surface as that Same at 2000 mag. shown in Figure 16. after etching in nital. 100 mag. Figure 20. Structure present after grinding off the heat etched surface shown in Figure 16. 100 meg. Shows grain refinement. owl» .99 a; r.— r... F... . s . . t . I e P.) In rd“. no) mo...- nu. . a u. r l. on” a I a a .5 .0. . r; ...... L \i. ‘0- '. v. ..r. 5.. .OJ ‘. .I re “‘5‘ e- 30¢ " rm. 7. on...“ w! .I. we... _ . .u c l‘ ’ . .... .. fl)... .rL ‘5 a A re. ....L. a. .... n r Figure 21. Surface of a coarse grained hypo-eutectoid steel after cooling from 1500°F. Etched in nital. 18001mag. r 1 ‘ 1.. Figure 22. Surface of a hypo-eutectoid steel etched after cooling from 11700F. Thé grains were visible at this temp- erature. 100 mag. Figure 23. Same as above at 800 diameters. Practically all‘the pearlite ‘ has been removed from.the immediate surface. . ’ . ..‘h l‘ ‘ Q. . s 1‘ ‘ . I’ t I . I ‘ .. --, .‘w ‘ z . e ' - > ’ ’ ‘ '4' 1 - . ' I -‘ ‘ , ‘ I ’ ‘ ‘ ... - I. i ..J V D ‘7; ' . . V . . , . . ' 7 ., . l “(J | ‘ "I -> ”:3 ‘l ‘ I ~ .._ . L‘8 4 - -... A“ .c' . 1 '1 ' ’3 C. ' v. - n 1‘ On. ‘ ‘. . l -' _ __._#w——-* {“25 f ..Y. -’ ' . _ , .\‘ ‘_ w .. ' a el ‘ . .. s a}. ~ recrystallization occurs. The temperature at which volatil- ization began was not investigated beyond this point. The following discussion describes the character of a film that occassionally occurred on a sample of steel heated and cooled in a vacuum. The nature of the film.was not de- termined due to the limited amount of time available. An etched sample of coarse grained steel was used. It was heated to 950° Fahr. and cooled to 4009 Fahr. The sur- face remained perfectly clear at the upper temperature but started to coat over with a very thin film.when cooled to 650° Fahr. Slight traces of brown film were first seen in various spots over the surface, then the ferrite between the grains of pearlite turned a dark brown; finally at 400° Fahr. the entire surface was covered with a transparent blue film, however the original Structure was still plainly visible underneath the film. When the coated sample was cooled from 11000 Fahr. the film.took on a purplish tinge which turned to a decided brown color when cooled from 12000 Fahr. The sample was then reheated to a still higher temperature. At 14500 Fahr. the film.started to break up; light spots occurred over the surface which grew in size until the entire surface was brilliant at 1850° Fahr. This was the maximum.temperature obtained and the structure was the same as that shown in the photomicrographs. (Compare with Figure 24.) Figures 25. 26, and 27 show the same surface after cooling at 100. 800. and 2000 diameters respectively. Figure 28 shows the surface after etching in nital. The important thing to note in this 11. Figure 24. Photomicrograph showing the structure of a hypo- euteetoid steel at 1825°F. 100 mg. Figure 26. Same as above at 800 mag. Figure 28. Figure 25. Same as above after cOoling from 18500F. 100 mag. Figure 27. Same as above at 2000 mag. Same after etching in nital. I... .... l. . ... 2/» a. 0V o e \r . .Mx . a. .. l 1.. e. u. . o , I L J . . . a u v - .4 Q, J ! ... at. . . ..-. a... ., I- l. 7 n... l v: ‘ . ‘ V . . . 7 ...-m ‘ — i v . . J ‘. t ' 0 . .... a; '10 n. '4‘ ..H.. C. _ . u- r .I\—' . , . , ., . . or.» Q. 2 . l u A .. . .. . . , s. . . .... .. v... .- ~ .- .IH series is that the pearlite has been almost completely re- moved from the surface leaving a ferrite layer which has a velvety appearance. The entire sample had a very brilliant luster. It looked similar to a nickel plated surface before buffing. The general character of the structure is the same as that obtained on samples that did not film. The author did not investigate this peculiarity, which for the most part had a tendency to occur when samples were cooled fromta temp- erature between 850 and 1000° Fahr. Some evidence was gained which indicated that it was due to various materials being volatilized from the specimen. The film was very pronounced 'when the insulation of the heating element extended above the polished surface, indicating that the vapors coming off the steel were trapped and could not get away from.the speci- men. Very little filming was noticed when the specimen ex- tended above the insulation because the pump could pull the vapors away as fast as they were liberated. The above phenom- ena have been mentioned in a hope that future work may lend ‘ some explanation. Figures 29 to 33 show a series of photomicrographs taken from.an unetched sample of annealed hypo-eutectoid steel. They are a record of the structure which was visible at the following temperatures respectively: 12700 Fahr.. 1370° Fahr., 1550° Fahr., looo° Fahr. (cooling), and 70° Fahr. (after cooling). When the structure first became visible it was very faint and did not show much contrast. As the temperature was increased to some 14000 Fahr. the grains became more 12. Figure 29. 'Unetched sample of hypo- euteotoid steel photoed at 127003. 100 mag. Figure 31. Same as above at l550°F. 100 mag. Figure 30. Sameoas above at 1370 F. 100 mag. Figure 32. Same as above on cooling from 155003. to looo°r. 100 mag. Figure 53. Structure after cooling to room.temperature. 100 mag. 4., .g' s - ¢. \" 1 ‘ l n \ \ 0\‘\ v ‘65 x a e A ~ d U I?“ L: T .A.‘. a} I" m 7w '.a' ‘ ‘ 3’ ”“9. f! 7-5 v4 ‘O. 9-» —?- 9.2L ion) 6 2.." _.-. “i: ...............a. _ . . . d.L...mvnW..\._ .... .euufiaflrv .... . .... .".:‘\< V ...s. K .\ ...... _. 3...... . .... .5. . .... . . .m...?.:....... ...R........\ Xyexwmwwx I ..w. n .N‘. .. \... \. .... n. x ‘43.: .... .... “dawn“ ......W... ......M... D ..- . ...? . . r. .< .\ O. ... distinct and were much easier to photograph. A further increase in temperature up to 1550° Fahr. did not change the grain formation but did cause considerable widening of the grain boundaries. The formation present at this elevated temperature continued to persist after cooling to room.temp- erature; however, the entire surface seemed to be slightly roughened. After cooling an oil emersion objective revealed that the pearlite had been removed from the surface as before stated. Figure 34 shows a photomicrograph of an unetched sample of coarse grained low carbon steel at 1150° Fahr. The normal structure of the sample before heating is shown in Figure 35. Without going into any more detail, it can be said that. as far as the microscope is concerned. the reaction of a low carbon steel at the critical is the same as that of a.mediwm carbon steel. Some observations indicated that the network became visible at a little lower temperature in low carbon specimens. Figures 36 and 3? show the structure of a hyper-eutectoid steel annealed to produce large grains of lamellar pearlite with the excess cementite precipitated in the grain boundaries. Figures 38 to 41 inclusive show the unetched surface of the above sample at 13750 Fahr., 14759 Fahr., 1600° Fahr., and 11500 Fahr. (cooling) respectively. The coarse network (Figure 38) outlines the original grains. The cementite in the boundaries darkened, thus outlining the grains more dis- tinctly at lOOO° Fahr. The first signs of the finer network 13. Fig‘lre 54 o Uhetched surface of a low carbon steel photoed at 1150°F. 100 mag. Figure 350 Original structure before in a vacuum. 1.00 mag. Figure 56. Annealed high carbon steel before heating in a.vaouunw Figure 3 70 2000 mag. /'.Jo\/’ . ' // , V'N ‘ I c. . .u\ .\\I\‘) u ‘ . I {72 1 -:\\ V I Figure 38. Unetched sample of high carbon steel photoed at 1375°r. Original structure shown in Figures 36 and 3?. 100 mag. Figure 40. 8am; photoed at 159003. 100 mg. Figure 39. Same, photoed at 14750Fo 1.00 mag. Figure 41. Same agter cooling from 1600 I. to 1150 F 100 mag. "' 7".11 5 ‘uw 5 "C4 ' V . no ._ .....L..7 7...... .. 3%,?“ WWW. . occurred at 11600 Fahr. The grains became clear at 1375° Fahr. They were much finer than the original structure and independ- ent of it. This new grain structure became more visible as the temperature was increased. Along with this change the boundaries of the original grains grew fainter as can be seen in the photomicrograph at 14750 Fahr. At 15750 Fahr. the original cementite boundaries were entirely gone and the sur- face consisted of a fine grained structure that looked like ferrite. The boundaries no longer looked like those on etched specimens but were spotted, indicating that a reaction had taken place which was not visible in the interior of the grains. Another sample of hyper-eutectoid steel was etched be- fore heating in a vacuum. The microstructure showed large grains of pearlite with excess cementite in the boundaries as shown in Figures 36 and 37. Figure 42 shows this same surface after heating to 14000 Fahr. The effect of the etch has apparently been eliminated by the recrystallization and removal of the carbon from the surface. At 2000 magnifications the entire surface was brilliant and looked like pure ferrite. The boundaries of the grains contained many extremely small spheroids which appeared dark. A.very light etch on this surface ate through portions of this ferrite layer and brought out a structure corresponding to the lamellar pearlite present before heating (Figure 43). Two distinct sets of grain bound- aries were present before and after this etch. The coarse boundary outlining the large grains is shown in the previous figure. The fine network winding across this grain is evidence 14. Figure 42. Structure of an etched sample of high carbon steel after cooling from 144OOF. in a vacuum. Original structure shown in Figures 36 and 37. 100 mag. Figure 43. - Same surface after etching in nital. 2000 mag. in, . 7 .») 9 a b 4‘”; .1 a '0 . . It .3 r. ...I. .. . .. I A rm. . e O ... a. .. u 4-. . . ..-.. L s. . ..7 .. i .0 a- ... ‘- ’I§ O 2. v’f' .ILA )5 ‘0. V! w of another volume change independent of the previously formed grains.- The fact that the etchant revealed the presence of lamellar pearlite beneath the surface showed that the ferrite layer was very thin. Traces of this thin layer are still visible as a translucent film over the surface. It is easy to see that a quick heat on a high carbon steel would not be sufficient to cause decarburization to a depth comp parable with that shown in previous photomicrographs. 15. Figure 44. Cooling Curves. [C i .1! [fi’m’ (/3! TI a. J [YIN/Lb ( t'i‘l’t/ '- I .7,“ )fi'.’ .1 £07111» 714/ A Wye/7,707.; M75. JV A T176574. 27h Egan/gr \ \ \ C \ \A \ \ \ 11.751.442- ‘\ C‘URV/ '7 A ‘rfi‘ - ”Y/D-[Vf/Cfb/l) .S' [/17 CU/I’VA'S' C‘Yfi — IVY/Fill” lfl/IT/K/f) {7/11 - - - COOZID ”(A VACUUM K/II/I) — (6’01/7’47/1 fMflf/V/f 2W5 PfiTSSU/ff Plvu I a i 41 A L i I 1 l t ‘L. /' _ 6 m [5 a) 2 5' INTERPRETATION AND DISCUSSION OF RESTLTS When a piece of polished steel is heated in a vacuum certain changes occur on the surface which are a record of various reactions taking place within the metal. The same reaction is visible on unetched surfaces as is visible on etched surfaces. An unetched sample of steel takes on a.matted appear- ance at about 800° Fahr. indicating that some grains are expanding more than others. As the temperature is increased very fine black lines occur over the surface which continually become more numerous, and at 1150 to 12000 Fahr. form.the boundaries of an entirely new grain formation. The entire surface is very brilliant and looks like pure ferrite. Time-' temperature data showed this reaction was not confined to the polished surface. The curves shown in Figure 44 were plotted from the following data: THE-TEMPERATURE DATA FOR A HYPO-EUTECTOID STEEL See Figures 5 & 6. 15 sec. Cooled in Air Cooled in Vacuwm Time Intervals Temp. Fahr.° Temp. Fahr.° l. 1500 1590 2. 1490 1495 16. TIME-TEMPERATURE DATA CONTINUED 15 sec. Cooled in Air Cooled in Vacuum Time Intervals Temp. Fahr.° Temp. Fahr.° 3. 1479 1400 4. 1465 1320 5. 1450 1250 6. 1440 1200 7. 1428 1155 8. 1413 1125 9. 1401 1125 10. 1391 1130 11. 1381 1120 12. 1370 1090 13. 1361 1050 14. 1351 1010 15. 1341 980 16. 1335 950 17. 1322 915 18. 1318 890 19. 1309 860 20. 1299 840 21. 1290 815 22. 1281 795 23. 1278 24. 1270 17. 15 sec. Time Intervals 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 391 40. 41. 42. 43. 44. 4'5. 46. 4?. CONTINUED 1266 1260 1258 1254 1250 1248 1244 1240 1234 1229 1226 1220 1218 1212 1212 1219 1221 1224 1224 1224 1222 1220 1215 TIME TEMPERATURE DATA Cooled in Air Temp. Fahr.° 18. Cooled in vacuum Temp. Fahr.° 15 see. Time Intervals 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. CONTINUED 1215 1210 1205 1200 1192 1184 1178 1168 1160 1150 1141 1139 1150 1121 1118 1110 1103 1098 1090 TIKE TEMPERATURE DATA Cooled in Air T3311). Fahr.° 19. Cooled in Vacuum Temp. Fahr.° TIME-TEEPERATVREiDATA FOR A HYPER-EUTECTOID STEEL See Figures 36 & 37. Cooled in Vacuum Temp. Fahr.° 15 sec. Cooled in Air Time Intervals Temp. Fahr.° 1. 1600 1570 2. 1585 1515 3. 1575 1475 4. 1560 1420 5. 1550 1365 6. 1530 1320 7. 1520 1275 8. 1505 1235 9. 1490 1200 10. 1475 1165 11. 1460 1150 12. 1450 1150 13. 1440 1150 14. 1430 1145 15. 1420 1130 16. 1410 1110 17. 1400 1090 18. 1390 1060 19. 1380 1035 20. 1370 1010 21. 1360 985 20. TIME-TEMPERATURE DATA CONTINUED 15 sec. Cooled in Air Cooled in Vacuum Tums Intervals Temp. Fahr.° Temp. Fahr.° 22. 1355 960 23. 1350 945 24. 1340 925 25. 1330 26. 1325 27. 1320 28. 1310 29. 1305 30. 1300 31. 1290 32. ' 1290 33. 1290 34. 1300 35. 1300 36. 1302 37. 1302 38. 1302 39. 1302 40. 1300 41. 1300 42. 1300 43. 1300 44. 1300 21. TIME-TEMPERATURE DATA C ONTIm 15 sec. Cooled in Air Cooled in Vacuum Time Intervals Temp. Fahr.° Temp. Fahr.° 45. 1295 46. 1295 47. 1290 48. 1285 49. 1280 50. 1270 51._ 1250 52. 1250 53. 1245 54. 1235 55. 1230 56. 1220 57. 1215 58. 1210 59. 1200 60. 1195 61. 1190 62. 1185 66. 1160 70 1145 74. 1135 The data was obtained by taking temperature readings every 15 seconds on samples of steel which were cooled first 22. in a vacuum and then at atmospheric pressure. The curves show the A1 transformation of hypo-eutectoid and hyper- .eutectoid steels to occur some 90 to 1400 Fahr. lower in a vacuum.than at atmospheric pressure. This being the case. the recrystallization visible on the surface of the specimen at 11000 Fahr. is a record of a reaction taking place through- out the metal. . What is this reaction? To answer this question we will consider the phenomena from.the equilibrium standpoint. When a piece of annealed steel is heated to the A1 transfor- mation the a1pha~iron in the eutectoid material changes to gamma-iron, which is more dense, due to the increased number of atoms in the iron lattice. This change is accompanied with an absorption of heat which mmnentarily cools the metal and causes it to contract. This seems to account for the different focal points which were described on page 9; hows ever. it does not account for the fine dark lihes outlining the new grain formation which is visible. This same network was visible after cooling from.the A1 transformation. The entire surface looked like pure ferrite regardless of whether the sample was etched or not. At high magnifications the surface of the grains had a velvety appearance. The boundaries were wide and many times contained a mass of very small globules or spheroids. Some heats showed these spheroids to be scattered over the inter- ior of the grains. Apparently something had been removed from the sample while at heat in the vacuum. A light etch 23. brought out a peculiar structure over the entire surface. The laminated condition of the pearlite was usually gone but traces of cementite were present in the form of globules very similar in appearance to the material present in the boundaries before etching. Apparently carbon had left the surface even at this low temperature. This same grain formation persisted when specimens were heated to higher temperatures in the furnace. The only change that could be seen was a swelling of the grains and a widen- ing of the boundaries; however. those specimens cooled from temperature well above the A1 transformation gave evidence of a secondary recrystallization when examined at higher mag-- nifications. This secondary network looked very similar to the one heretofore mentioned except that the boundaries out- lining the grains were much finer than those outlining the volume change that occurred on heating through the A1 critical. It had a.different point of focus and was perfectly independ- ent of the wide boundary network which showed it to be another volume change resulting from some sort of transformation. Specimens cooled from within the A1 transformation showed no evidence of this volume change while those cooled from.higher temperatures had a double network present over the surface. It seems probable that it is a record of the transformation which occurred at the upper critical. Since the allotrOpic transformation of alpha-iron to gamma-iron takes place grad- ually when a piece of steel is heated above the A1 transfor- mation, it does not seem that the volume change would be 24. sudden enough to leave a record on the surface; however, the fine netowrk can be explained by studying the reactions that take place when a solid solution of austentite is cooled through.the critical ranges. It is a well known fact that critical temperatures obtained from cooling data are consider- ably below those obtained from heating data. This indicates that there is a lag due to a hysteresis in the metal. The trans- formation of gamma-iron to alpha-iron does not occur in a mass of steel until the alloy has cooled to some temperature below the normal critical. Then suddenly a considerable amount of the excess constituent(ferrite or cementite as the case may be) passes through the allotrOpic transformation. This is accompanied by an evolution of heat which causes the mass to increase in temperature and conseguently expand a.small amount. Probably the fine grain formation present on the surface of a piece of steel, after cooling from above the critical in a.vacuum. is a record of a change in the metal which occurred while cooling through the upper transformation range. This also accounts for the fact that the boundaries are not as wide in the fine grained network because they were not present at the elevated temperature a sufficient length of time to allow for much of the ferrite to volatilize. Another reaction was observed simultaneously with the volume changes discussed above. It has been shown that speci- mens etched before heating lost the identity of the etch with the occurrence of a volume change at the lower critical temperature. It has also been shown that specimens etched 25. after heating in a vacuum gave evidence that most of the cementite originally present in the pearlite had left the surface. This phenomenon has been referred to as volatili— zation from the surface. To further verify these statements another specimen of hypo-eutectoid steel was heated to a temperature well above the upper critical. After cooling the heat etched surface was covered with a thich.nickel plating to preserve the existing structure. Figure 45 shows a section taken perpendicular to the heat etched surface. The surface, which originally was smooth, is now very irregu- lar. The indentations are evidence of volume changes having occurred in the metal. It also shows that the pearlite has been removed from the immediate surface leaving a ferrite layer consisting of relatively large grains. Rawdon and Scott have reported similar results and concluded that cementite had volatilized from the surface. The depth of this carbon- 1ess layer in many cases was to great to be interpreted as evidence of recrystallization in the filmed surface layer produced during polishing as advanced by Sauerwald. Schultz. and Jackwirth. It is very evident that the depth of the decarburized layer is dependent on the rate of heating and the1maximum temperature obtained. The evaluation of this relationship gives a fertile field for investigation. Rawdon and Scott concluded that the depth of the carbonless layer became less pronounced the higher the temperature was in- creased above the Al transformation. They attributed this to the increased rate of diffusion of carbon from the interior 26. Figure 45. Section perpendicular to a heat etched surface. 2500 mag. Nickel was plated on the surface to preserve the existing structure. The roughened surface and decarburized layer are plainly visible. LA. . 2‘ . .l ' ". ~. ‘v n 51 L 5 ll .«. -. c . .« .5 (. ‘a ..a- 5" V' . .“ 1 v ~' \VI"‘ 1 H ~0- . r' x I‘ u 1. . .. Y" 1 " ~ ' ( -1. a. ; . .' =7 ‘ ff 7 ;_ V . 4. 4‘! ... - . ‘x' - 1' -'| ' 7A I \ (g o -. kw i 1 $- 4'. .3 . . ( .-' ’ ..." ' .7 U I\ . ' ..., , l A I "’ . ' .. .3 _- f 3 "I: " -.'. v" 9 ‘ . ‘ — .A ,. l ' I ' , ‘ - .1 O H A; - t.) 7 7' a r. _- l 4. > ... 1' ‘ ‘ O a .. . . I .’ .. u ’ W ‘\ .—‘ ) \.’ 4". 1 . ff- . c J -a/ 0 a a ‘- .13 ii 9+ e 7 . ‘ I I \,'e 7 ' I . . g—a A of the metal to the surface thus masking the effect of decar- burization in the surface layer. A The exact nature of this volatilization is not known.‘ Without a doubt carbon is not the only volatile material in a piece of steel. The widening of the grain boundaries at elevated temperatures is probably due to the volatilization of small amounts of ferrite. Impurities and dissolved gases may also vaporize at elevated temperatures. Granting that this is true. observations revealed that most of the volatilization took place at the grain boundaries. The ferrite layer present on samples after cooling in a vacuum was much rougher on samples etched before heating. indicating that very little ferrite was volatilized from the face of the grain. The fact that the dark material was concentrated be- tween the grains instead of in their interior was evidence that changes were more pronounced at the boundaries. Rawdon and Scott reported similar conclusions on the persistence of slip bands in c0pper after heating. showing that very little metal was leaving the surface. The dark lines outlining the grains on heat etched samples are produced in two ways. Small particles of iron are volatilizing from the edges of the grain thereby roughen- ing them and causing an absorption of light instead of reflect- ion. Particles of cementite, carbon, or other volatile materials are also lodged in the boundaries and appear as small spheres. The volatilization of cementite has been spoken of rather freely in the above paragraphs but no attempt has been 27. made to explain its physical state at the various temperatures, which may lend some eXplanation for the observed phenomena. Cementite in normal pearlite or as the excess constituent in hyper-eutectoid steel is made up of iron and carbon chemically or physically bound together. If we assume. as some do, that iron below the A1 critical exists as a body-centered lattice, called alpha-iron, chemically bound to the carbon, and represent it by the formula Fegc, it will be necessary to eXplain why some alpha-iron is capable of forming a solid solution con- taining not over 0.04 percent carbon, while other crystals of the same lattice formation can combine chemically with carbon forming a compound containing 6.67 percent carbon. This in itself says the properties of iron below the A1 criti— cal are not the same in the two constituents present. If the iron in cementite is of the body-centered lattice forma- tion there must be changes in the atoms themselves in order to account for the difference in their affinity for carbon. The very fact that carbon is chemically bound to the iron molecule destroys the ordinary conception of the simple body- centered lattice characteristic of alpha-iron. The existance of carbon atoms in the lattice would alter the atomic forces and thus change the preperties of the mass. Then too. if we consider cementite as a chemical compound it must have a definite crystalline form in order to follow certain funde- mental chemical laws. Its prOperties would always be the same regardless of the heat treatment given providing the temperature of the treatment was below the decomposition temp- erature of the compound. If the temperature of the treatment 28. was above this critical value the cementite would be destroyed. Then if the rate of cooling was changed from that of the pre- ceding treatment it would be very possible to change the man- ner of formation of the new constituent and produce a material having entirely different properties. This new constituent would have a different crystalline form and should not be given the name of cementite. Let us consider the other possibility that cementite is a name given to a constituent of steel and is made up of carbon holding varying amounts of iron in solution. Possibly on first thought this conception is a little ambiguous. But why should it be? Chemists have found that all substances are at least partially soluble in the liquid state. also,“ practically all metals are soluble in each other in varying amounts in the solid state. Why then should we say that car- bon is soluble in iron in varying amounts, then. turn around and say that iron is not soluble in carbon but is capable of forming a chemical compound the formula for which does not satisfy our present conception of valency laws? Since cementite is so susceptible to changes in prOperties by heat treatment it would be more logical to consider it a solid solution of carbon and iron, the preperties and stability of which are a function of the temperature? Let us see how this conception can be used to explain the volatilization of carbon from a steel when heated in a vacuum. When a hypo-eutectoid steel in the annealed condi- tion is heated through the A1 critical the iron in the alpha- ferrite changes to gamma-iron which is capable of holding a 29. considerable amount of carbon in solution. Immediately the carbon in the cementite starts to diffuse into the gamma-iron and form a solid solution of gamma-iron and carbon. Probably cementite is very unstable at temperatures somewhat below this allotrOpic transformation thus augmenting this diffusion, the same as any other solution becomes very unstable as its decomposition temperature is approached. In fact it would seem that since cementite can be precipitated in so many ways by varying the manner of cooling. it is-probably continually changing its physical preperties as the temperature is increased from that of standard conditions. Without a doubt this change would be accentuated by heating the steel in a'vacuums The filming of a polished surface at temperatures below the trans- formation range may be due to the unstable condition of the cementite at these temperatures. Possibly some carbon is re- leased from the cementite due to its unstable condition. Since this reaction was noticeable at relatively low tempera- tures, the carbon released could not be taken up by the iron mass as another solid solution because the alpha-ferrite is already saturated. This would leave minute amounts of carbon free on the surface which might appear as a thin coating. If we think of the allotrOpic change occurring at the A1 critical as requiring some time. there is a period for each crystal grain when the carbon present is not held in.a stable condition; that is. the carbon on the immediate sur- face would be free to leave in atomic or molecular fonm. Removal of carbon in this manner might be sufficient to cause 30. etched pearlite to loose its identity even before the new grains were well formed. Immediately following the allotrOpic change of the iron in the critical the carbon starts to diffuse into the newly formed gamma-iron lattice and if the temperature is increased above the 51 critical the excess alpha-ferrite in hypo-eutec- toid steel likewise passes through an allotrOpic transformation, thereby augmenting the diffusion of the carbon. An increase in heat also increases the rate of diffusion due to the in- crease in the thermal agitation of the atoms and molecules comprising the metallic mass. As a result, carbon continually moves to the surface and is removed through the grain bound- aries. ‘When a.specimen is cooled from within or above the critical range very small particles of the carbon remain trap- ped in the boundaries and appear as a mass of small spheroids when viewed at high.magnifications. It.may be that impurities and gases in the steel act in a similar manner and also add to the formation shown in the previous photomicrographs. Since volatilization is continually occurring it is evident that the composition of the surface material is continually changing and that the observed recrystallization within the critical range is not characteristic of the entire specimen. As soon as the steel is heated through the A1 critical the carbon content of the immediate surface is de- creased and the observed phenomena occurring thereafter are characteristic of the ferrite layer and not necessarily the entire specimen. This accounts for the single network 31. present on samples cooled from temperatures within the A1 critical since ferrite has only one allotrOpic change. Con- sequently the grain formation produced on heating is not altered in cooling through the same temperature interval. Likewise the secondary grain formation. produced on samples cooled from temperatures within the as critical range, is a record of the gammapalpha transfbrmation in the ferrite layer. As before stated the record of this allotr0pic change in the ferrite is much.more pronounced on cooling than on heat- ing because of the suddenness with which it occurs. The absence of spheroids in the boundaries of the fine network is further evidence that there was a change in the ferrite layer and that the amount of volatilization through the boundaries had been reduced to a minimum. 32. SWTHARY Ale CONCLUSIONS 1. When a steel is heated under a reduced pressure to temperatures within the critical range evidence of a recry- stallization becomes visible as new grains appear over the surface. These grains are formed as a result of a volume change; however, this change did not occur as a wave over the surface as shown in the moving picture film taken by Rogers and Wert. 2. Specimens of steel heated to temperatures above the upper critical showed the presence of two grain forma— tions, one superimposed upon the other. They have different points of focus and were a record of two distinct volume changes that occurred in the metal. 3. Recrystallization occurred at a lower temperature when steel was heated and cooled in a vacuum than when heated and cooled at atmospheric pressure. 4. Carbon volatilizes very rapidly from.the polished surface of a steel when heated to a temperature within or above the lower critical range inia vacuum. 5. Vblatilization is more pronounced at the grain boundaries than over the face of the grain. 6. It is not possible to see cementite in pearlite or the excess constituent in high carbon steels go into solution with gammapiron in a vacuum. 7. The ferrite layer left on the surface of a piece 33. of steel after heating in a vacuum is not a film but is a crystalline mass. It extends an appreciable distance below the surface. 8. The same reaction is visible on a sample of unetched steel as is visible on an etched specimen. 34. ROOM USE WU" I. t I, :43, \ g. ‘0‘... . ' I'III'- l, s .d ,. ‘ ’ffly‘l L‘ ".I 'q‘vr' ' ) K k « ° 5 ~ “ " l "I A, . . f . ‘ . ~ ' \ ‘fi‘htx Iv?" ‘3‘“ I 'fi"~ “" I' ‘ J k ‘| I ' - "“ .- ~.. '.' [a ' . ‘.' , '.‘ ‘ *t H. “I"..J f‘ ? &'”‘|'~¢ ‘4 ‘I ‘1 ‘ (‘.’.O'. h. ‘J'..-l'~‘ 3 i ' . "94“” my. “WEI-"o .z-u- Wig? AOL-"L '1‘ I, \ . - s ‘ I. . .I ".1‘ ‘. ‘.‘- .I’I. '.‘. l 1' . ' v p I g.” ‘.fij‘w‘f‘. in“ ".o’i 4"“ .. .‘ . .~ ‘ -- ‘ . - - . » .e . *LJ ,I a fi-( ‘ 1VQ t'J' Q I 5v;§1" .°!?. ‘I ‘-' . - I, - rn " H v .. . ‘ I . . I .. " a ‘ ..-:\’--. -}».“"‘-‘-‘u§ 5‘; :{IK " 'I - \ l '. 4‘ " . r ' ‘ u . 'I'w I I, , ' ' has; Mfg; M Jews I: In .' . --,. . ; -- Mia's-fl m ' ”Jean-WV 4-; -.. n 1 , . ~ - :- U.’ s ‘ 2; ‘; . ‘.' ‘5. ’1‘. ‘ ‘- 1‘ ‘\' s 1 ‘.’ ‘.‘t‘ ’1‘ . 13"}1’"I¢I \.- '.'I’ ' ,' M :II" I". "‘l‘“‘ -. .' I ' I . I f I .II . . . '. . . II '4) ”’5‘ f' ‘.‘“ ‘.‘ £33K» y’ I" .}% 3“. "b . f" {v ' ‘1'. *3 ' . ,' . ‘ H ~' ‘~ . .‘ 5 r ‘33s 5 31‘" ,- " *J‘wt‘fsh'fih“ f" ‘2‘":- ‘x ' V .’I " ‘ “i, Fag-‘.‘“ " W 1‘7"}, '»* -’ gs ,"\‘ '.q 9 1 {‘.‘-W 5-.“ '4';_'h"_.' ' '324 1‘0 Two ,-S- ' ’.‘." . .‘ .(MI, s .r‘ '0 .‘1 «H. 0": \ 7 \‘£F§"‘L\.‘JA ‘. I ' ‘ . )‘f.€‘ Ix " I =’ I l "1." ’ a .v 0 7‘. .' " _v ~ . . I u_' I i‘ hf ? ’t: ,1" - . . , ': ‘.' I'. I- ' .0 s . a t . .r .3‘ t§ffi ‘ ,I\ -I\ . . . 0 t... . . I'.I - f .' .... ‘5‘} . . . '. . d \' >5- 1 ) ‘g:'; t _ .I N); . ‘.‘! 3’“ .fi .' .‘ ' $ , . 4 ‘ f f‘{‘ . ‘. ‘ .— I' '* "is: ,l I I. 1.. A . ' r ‘.'-o- ‘ . 42L" '1 {:3 3‘ ' - ~ $ e f o .v 1‘3- ~ ‘ \F " \fi ' ' ' ‘I ' ' ’r w ‘-v.'.cr5r&°—-.‘*'3-* . ' I il‘v‘fdk- '. n; -. “‘4‘ ‘.' s" L ‘1 ,V‘.'? ‘ ..- 3 5931-, .. ‘.IJt :. K ‘ , .I .. 71’. Ila! . F; .- .-,., III-Ete‘oR-x’t‘yyHI 13.1. .. N . t , "'.. ‘I . ’é T. ‘0 ”T '. « "~ 31¢". ‘ "IV-(Tr 1‘11..-' . w . V -. I Q' It. :' ”‘.‘" I‘%;- ‘\&'3‘ ‘ y " . ‘1 ’7'» H ’ "h- t'-'--‘ It. 6‘ f" - .1 3" .~ WV ' 'C'.‘ '.‘;3 . ‘3'};1’6". «e . \’ ,t 5;~.';I.;a_oIn-.._I.. ,Is‘ £3 .I. t , ,2 y» l 1": v ‘.'LLJ'bfim‘H I “M' ' ' -. - . . _ 3K in ‘.‘..r‘hi‘ra}? “*h "3‘ ' ‘ \ V L5" ' ‘l ‘9‘ .1 ° AI , , _- ‘I r _ ' . I ' M . tr‘e )t'l ‘; “.‘fi I“ “1;: ‘W _,.I I, fl?“ ‘.‘" '., _ . . .. I}. II II I . ~ . . . _ I I I .. . . .I I .‘I. I II I ‘.‘; :IIIIII..I fut? “fi‘fi‘lL-‘i I \I' . .31... ,._ ‘ . s- .)‘ . . ._ ‘. I I ”g; I? .‘I I ‘.I. .I I ... "V‘ ‘2‘ .'~ *1": N "I. ‘ H , J .1 ,-’u- . , _ . ~ - ‘. I K4I~¢~o>§j I . . i t ‘5' d I‘ III:{’:;‘ ... y ..., I -. . I ' ‘.’“, _o ‘.g" I , ‘ .' u . . fl. .- .f’.‘ .‘-_. '. .", a o‘ ‘.‘. l( .‘! . - ‘ A.. I O ‘1}. ’3‘": ‘ , ‘3: j; ' "L"! ' ‘.‘ . I ' ' ‘l h In ‘.‘ J " ‘l'. ' . _. t ‘ >2" Ix’kzet“. \‘ I i"... _:". ..- . |-'.' ~‘ I'II fl k -‘I» . -. I .‘ . . . - II ‘ . .I I,' . I- I II.I‘.I .‘I 'I.I ’:'.Z"'-‘_‘ I I I '5 -L. . ."k . . . _r I I I. ‘ Ir _ q} ‘\. .‘ l"; ‘~ 1 ‘ . I ‘4' I- . I , . '.‘..- ‘ o _ . I _. . .' I We ...-1; .. 1.. x ~ -. we a. s»: - « , . . ; . .-.. . _t. ., . I . .‘u. . L) . k v - . . ‘1 . . . - .. - r'- . 7‘s: ' I I: ‘5‘ aIII‘ia'I‘. . M In. ‘ 3.1%."! L ‘555, . ‘i‘f'i-I'OF'Nz.‘ I‘- ,5 ‘.' “x 'I' - , J ‘ _‘ Y “ . _ ' f v ‘ 3.175.; :I‘I‘WI'J' , . . : :.-~s,s_:~,:,,;,,«m 2“” :3 any; a: s __. -. . a , ;.. 4 __. I _ a. . l'c .‘ f?.u,?".’fs"‘,‘- :l:";¢‘h"litikklg .3...‘Ir'-&x. .- R- . .’ 5 . g :2 I». 3’." ;I 'I -AII . I III? r." ,I . I I I ~-\;-\i: . _ I? I. . I , I 1 ,1, ‘ I£.I.IIIT .I..I‘ Is . . 5‘ '1 d _ .- , I 'I- - . _ »~ '. . ‘ ."I I 4 . '.~ V . '. . . I - " .\ '. " .'. '.~-.- V. I o‘ I‘h'-'I_4 .I- ~ I . ' 5 . i . -‘ .‘ -r".~')’..*“ '- " t-. ‘ - .’ ' _ 3:. u " . : , “ .' ‘ ' . ‘- ‘ . ‘ " ‘E‘.”" a .p' ' 1 . . g '5... ‘ t h. .I ' I . I 'u i ' ‘ ' ‘s ,‘ I . 1' Q ' ' :f‘dankflt QR ‘1 iirf‘)'\a‘f ' ‘.' :'. F r. I I -" - . . t . ' .lq' ‘.‘ V ‘ I ‘ \I‘ ‘. . y <' _ \‘ ~' *1... “fit? I? ,~_'I‘II'\’:‘.I#. y-‘li ‘ lII‘ 1C .4: . ' II_'n. ‘ I . - I _ I_ ‘ . '. II . fi":.? 3‘ " .g' n .-.' .. ) ..¢:.3~:-7: -' -. an " .t up 7 -- I i" , _ .qI ,I .I .I I ‘.‘}, II'I.I‘III .\ .I . '.‘ff'afz' - ‘00.. | . . '.'. - . " ' ' . >I. A .' i . - I . I . ’ n r . I . a .- _ . v . . w , . \\ I I . '.‘ - _ . . . a . . . . II .I .‘f b. . ' 1" C ,‘J. . . ' I. ~ .' ' -- _ ‘ . . . . - a - I ‘ . - . 4 3 . I 1 J‘ ‘ . _ 7 ‘ ' e 1 , ' . , . .‘ V ‘.:y. ..I . , I - .- - v_ . i -. o. - - . ~ ’ . .' ’ .v - . , s ' v. > .' , ‘ ‘ ' ' ‘ ‘~- ‘ . . . U .I - '.‘ ‘ ‘ I .I 7" .¢l‘ ‘r’rk‘ '.'" ~ H V 4' ' r» z‘ : I. . r : :- . .. I , I I I 'l .1_' ' . . . . , ' I . u! I ," I i .‘ ' ' . .. . ... . ' , p .. . I , :, I‘ .. .. \fitt'l ‘ ‘ ‘ . ' .'-‘ . ' . ' ‘ - . . . ‘ > " 26"“ I r I 1 t“? I }£ I‘ r. .' I 1 " .‘ . ' n ‘ # ~ ‘ . s .I '. .- . \V' 'A . '. VII I ." ‘ ‘ '— | ‘ s I. ' ‘ A I. - '. .’ I o ‘ : 3 ‘ ‘ . . MICHIGAN STATE UNIVERSITY LIBRARIE I HLIII III II” IIIUIIIHIH