SCIENTIFIC ABSTRACT ROZENBERG, V. M. - ROZENBERG, V. R.

Document Type: 
Document Number (FOIA) /ESDN (CREST): 
CIA-RDP86-00513R001445610013-6
Release Decision: 
RIF
Original Classification: 
S
Document Page Count: 
100
Document Creation Date: 
November 2, 2016
Document Release Date: 
July 13, 2001
Sequence Number: 
13
Case Number: 
Publication Date: 
December 31, 1967
Content Type: 
SCIENTIFIC ABSTRACT
File: 
AttachmentSize
PDF icon CIA-RDP86-00513R001445610013-6.pdf5.41 MB
Body: 
SOV/126-6-2-19/34 Study of the Influence of Preliminary Deformation in the Cold. State on Creep inter-atomic bonds in the crystal lattice.' Thereby, it is necessary to take'into consideration that conservation of a structure which ensures a high strength at elevated.: temperaturns also depends on the strength of the inter- atomic bonds. The creep speed of specimens hardenea.by deformation in the cold state and notpreliminarily stabilised increases sharply in the temperature range where recrystallisation occurs. Thereby, the stage:. with a steady state creep speed shifts towards longer: time durations, Under@the simultaneous effects of temDerature and stresses, softening of preliminarily,. deformed specimens it more intensive than solely as a result of elevated temperatures. There are 5 figures, 4 tables -Tid 6 references, 3 of which are Soviet, 3 English. ASSOCIATION: TsNIIcheriaet SUBMITTED: December 17, 1956 Card 3/3 i. Ailoys--D,eformation 2. Alloys--Creep 3. Alloys-- Heat treatment 4. Alloys--Test results SOV/126-16-4,:;-32/34 AUTHORS; Rozenberg, V. M. and Gradova, L. V TITLE: Phenomena Associated with Grain Boundaries.in the Case of High Temperature Deformation of a Solid''Solutionof Titanium in Kickel,(Yavleniya, sv-yazannyye s I1@ranitsa,mi zeren pri vysokot6mperaturnoy deformatsii tverdogo, rastvora titana v nikele) PERIODICAL: Fizika Metallov i Dle-tallovedeniye, 1958, Vol 6.: Nr 4- pp 765-767,(USSR) @ABSTRACT: Data are givPn,relating to the displacement of, grai -n. boundaries in solid solutions of titanium containinp- 3.60% Ti. :The stretcqng of the specimens was effected in vacuum at 900 C after preliminary., or annealing at 12060C for 2 to 3 hours. The micro-structure- -c-he dis @th photos Figs 1 and 2., show place@ent of e ;zrain boundary in a sRecimen stretched at 700.0 by means of'a load of 8 k&/mm,--. The displacement of the bent boundarv on which the observations were carried out,took place in the direction away from the cent're.of,curvature, This seems to indicate that the moving force of the displace- ment is not the difference in the magni-mde,of the surfa Card 1/4 energy of the@gra ins but the difference in the elastic SOV/126-6-l'-':12/34 7 J Phenomena Associated with Grain Boundaries,in the Case a 4i-In Temperature Deformation of a Solid Solution of Titanium in Nickel energy of each of the grains, Fig 3 shows the case.of displacement of boundaries which is accompanied by @hel formation of @liding lines in the specimen,(Ni, 600 C,@ cy = 3.5 kg/mm ). At a higher test temperature. the . -displace'ment of boLuidaries can be. so, large that in, individual cases it is a grad al possible to observe, u cessation of the grain. It canbe seen from Fig.4. how an initially large grain is gradually "eaten up". by its neighbours. The photographs depict individual' stages of thedisplacement. It can be.seen clearly that this takes place with breaks: each break corresponds,, -to a certain position ofthe boundary. :The,observed change in the structure can also be explained Dy-.the displacement of the grains relative toeach other.:.Such a displacement should bring about a buckling of individual grains which, under the microscope,can be seen as the., Orowth of one grain at theexpense of another (this was pointed out by E. S. Yakovleva). Three photographs are Card 2/4 reproduced in:Fig 4 and it can be, stated that the , SOV/126- 6-4-32/34 Phenomena Associated with Grain Boundaries in the Case of@:High, Temperature Deformation 'of a Solid Solution of Titanium in Nickel observed picture is due mainly to the growth of.one, grain at the expense of the other and not to buckling. This view seems to be supported by the conservation of traces of original boundaries at 'he individual stages of growth and also by the possibility of simultaneous focussing of all grains durin@@ observation with a microscope and a maSnification of 200 times. It is.known. that the grain growth in the case of selectilye recrystallisation is stimulated by the difference, between the starface energies of -the conjuc@-ate grains, Selective rk.-irystallis.ation can take -olace at a sufficiently high, temperature when the mobility of the atoms becomes high enouF)- Since on the one hand, prior to applying 01 tens@,Ie stresses, the specimens were annealed at,1200 C and on the other hand, cases were observed@of displace- ment of the boundaries which lead to an increase in the grain surface, the assu@iptlon can be expressed thatin Card _3/1+ the here described cases the state of minimum free.energy S0V/l2c---6-4-q2/34: Phenomena Associated with Grain Boundaries,in the Case of HiFh Temperature Deformation of a Solid Solution of Titanium in Nickel to which the system tends during stretching is not baser@ on achieving the mininum surface-energy but on redistributing the elastic energy between the grains. The total magnitude ofthe elastic energy is determined. by the magnitude of the applied stress. There are 4 figures. (Note: This is acomplete translation except for the introductory paragraph) ASSOCIATION: Institut metallovedeniya i fiziki uie-111-allov TsNIIChM (Institute of Eetal@og;-?aphy and Metal Physics TsNIIChg) SUFvIITTED: May 7, 1957 Card 4/4 SOV/126-6-6'11/25 AUTHORS: Kaminskiy, E.Z. and RozenbeTgi V.M.__ TITILE . Study of the Influence of Dissolved Carbonon the Re^rystalli2ation and the Long--duration Strength of an 3 ' Ir@l--Nickell Manganese Solid Solution (Izue_-heniye.v1iyaniya rasivorannogo uglereda na rekristallizatsiyu _J dlitp.12nuyu pro,-ahnost! zhelezonikellmargantsevogo tverdogc rastvora) PERIODICAL: Fizika Metalloir i Metallovedeniye, 1958., Vol_6@ Nr 6, pp 1036 - 1039 (USSR) AB15TRACT. In th-,_ work dese--ribed in this pap r, the authors determined e the temperature cf the beginning of recrystallisation And they investigated the dependence of the long-duration strength of nickel--manganese steels on the concentrat-io'n of the dissolved carbon., the temperature and the st-Tesses. In the'i _n--restigated steels, the carbon was' in Ithe dissolved stalle and not in the form of carbides. All the investigated steels were austan-J.-tic and their compositions were as follows: Cardl/3 SOV/126-6-6-11/25 ."Study of the Influence of Dissolved Carbon on theRecrystallisationt and the Long-duration Strength of an Iron-Nickel-Manganese Solid solution Melt 0 Mn Ni Fe Heat Treat- Grain size, Nr % % % % ment after ball forging 588 0. 05 8-? 15-03 nst 1125 0C-2- 5hrs 6-5 589 0.4? 9-5 1---00 " 1100 0C-:-2 6-5 3450 0.95 10 1..@.00 " 1100 C-1-5 6"5 To exclude the influence of the grain size, each series of specimens was siabjected to a.heat treatment suchthat the grain size was the same in all.cases (see,table above). To eliminate.oxidation, specimens were tested in, argon atmosphere. The in-itial crystallisation temperature. was determined by the X-ray method after deformation of the specimens by W16. In Figure 1, the deDendence is graphed of the temPeratuxe of the initia 'l.crystallisation, of steels, on the 'concentration,of carbon in the solid solution. In Figure 2. the dependence is grqhed of the long-duration strength of steels onthe concentration of carbon in the solid solution and on the test.temperature. SOV/126-6-6-11/25 .Study of the Influ' so-red-Carbon on the Recrystallisation ence of Dis and the Long-duration Strength of an Iron-N-ickel-Manganese,Solid Solution In. Figure 3, the dependence is graphed of the long-duration strength of one steel on the test temperature. The following conclusions are arrived at.@ 1) latroduction of carbon into =on-nicke-31-manganese- solid solution reduces, the initial reary5tallisation temperature and thus indicates that the dissolved. -:.arbon ::@educes the resistance of the solid solution to the effe--ts of the temperatureg 2) the carbon dissolved in the solid solution brings about a-n increase in the breaking strength -a-Ii.relatively high defor- mation speeds and a deicrease in the strength at relatively. low deformation speedst the higher the 'lest temperatuxe, the higher will be the def6rmation speeds at which the advantages. of the -c@arbon-free solid solution will manifest themselves; 3) the here described influence of the carbon i's apparently- due to,a d - - e rease in the strength of the interatomic bonds and an increase in the distortion of thelattice of the solid solution. There are 3 figures, 11 table and 11, Soviet raferencis.- ASSOCIATIO TsN IChM D May 6. 195? 18(7) PWR I BOOK EXPLOITATION SOV/3355 Akademlya nauk 533R. Inatitut,m Ietallurgil. Haucbnyy *over o, p prablem* zharoprochnykh apl&vov 1smlftdOvAnIYa PO zharoprochnym aplav&m. t. IV (Studies on Heat-r,- xistant A.1joys, Vol, 4), Moscow, lzd_vo AN SSSR 1959 400 . , p, Irr'Lt& slip insert ad. 2,200 copies printed. z> Ed. of Publishing House: V. A. Klimov; Tech. Ed. . A. P. Gu3eva; Zd1tDrLAl -Board: 1. P. Brdln, Academtclan; 0. V. Kurdyumov /V , Atad"Iclan; X. V, Aseyev; corresponding Member, U=SIR Academy of L Sciences; I an I. A. Odin&, 1. 14. Fav ov, d 1. P. Zudln, Candidate or, Technical Sciences. PUN?= ; Th is book is intended for 0011, Ilurglata concerned with ''the structural metallurzy of alloys, COVEUCIR: This IN it collection. 0f specialized studies or various problems in the structural metallurgy of heat-re3letant alloys. Some ar& concerned with theoretical principles, now, With dnj. crIptions of new equipment and methods, others with p.-opertlea peciflad conditions are studied and reported on, For datftils, *a Table of, Contents. r The articl,63 are sccoopml4d by a num- : ber or ref4rended. both Soviet and non-Savlet. TABLE OF CONTENTS t 3av/3355@ UU41,49 (Cont.) and v. I., syutkina. mechanism Of High V`,k-_1-W3.,-Z-.-S kel- d Ni c an Temperature Copper Solid. Solutions QbAtk1ru_ and ans format ions A 3tudy or 3tructural Tr nyo xwrop 41 In Rs&t_ -Aluminum A11074 , : .rqsi tan opper and V. P. SukhovarOv. Con ;. V of e c5nWekkanlng Efr ee ceraiO4 t Stimuzating h In D#fOrEWtIO83 onahlp Betv..n D.ror%atlon In the I reep 1~11 During ant along tho Bound&- in Nickel on the Equivalence of thd Effect of Rate and 64 10 Of Plastic Plow _ -qemparnture f 3train on the rrocess V.,3_RIkhAl6nkOV`, and R. 0. PAdfttOVM- Detntyar Card 3/12. K_ sov/18o-59-2-9/34 AUT-HORS: Minkina Ye A Preobrazhenskaya, N.V., and-RozenbergV.M. (MOSCOW3 TITLE: Study of the Deformation of Nickel During Creep (Izucheniye deformatsii nikelya pri polzuchesti) PERIODICAL: Izvestiya Akademii Nauk SSSR, Otdeleniye tekhniches-kildh nauk, Metallurgiya i t6plivo, 1959,Nr 2, pp 48-55.(USSR) ABSTRACT: V.M. Rozenberg and L.V. Gradova (Ref,l) and. V.M. Rozenberg (Ref 2) have previously sho,@m that the effects found to occur in metals.and alloys during creep are applicable to the particular case of nickel..'Inthe present work the def ormation of grains, and .,that . due , to I. relative displacement of grains were investigated. The nickel used contained 0.02% C) 0.04% Mn, 0.006% S7 0.006% P7 0.08% Tij 0 .23% Fe and traces of Al,and Co. I Qualitative estimates of.deformationwere made by,the method of McLean (Refs 3M. Extension of,flat test,, pieces with surfaces prepared for observation was carried out in a vacuum installation (10-3 mm Hg). The measured- values of displacement along slip planes 9nd grain boundaries and the number of slip lines and grain Card 1/3 boundaries are tabulated, together with calculated values sov/18o-59-2-9/3)+ Study of the Deformation of Nickel Dur-ing.Creep. of the deformation. The measurements werecarried out at 400, 6oo, 675, 700 and 800 OC7 with stress values.of 13 - 2 kg/mm2. @Figs 1, 2 and 3 showtheaverage valueof displacement along visible slip planes, number of slip lines and value of displacement along grain boundaries7 respectively, as functions of time (hours) for 40.0 and 800 OC and stresses of 13 and 4 k /mm2 respectively, are shown in Fig 4. From the slope oi the line (Fig -5) of. logarithm of time to attain a,given displacement value versus reciprocal of absolute temperature an activation energy for inter-grain displacement for 600 to 800 OC and a stress of 4 kg/mm2 of 36 k.cal/mol was calculated-, Figs 6 and 7 show families of curves, for,400 and 800 OC, respectively, of total elongation and those due to slip within grains and at,grain boundaries vs. time for@ various stresses. Fig 8 shows the difference between the total deformation and that accountable to these two effects related to total deformation as functions of. time for 400 and 860 OC.. The relation between displace- Card 2/3 ment along grain boundaries and grain deformation for these two temperatures and various stresses is shown in_ sov/18o-59-2-9/3L@ ,Study of the Deformation of Nickel During Creep Fig 9 to be linear.. Figs 10 and 11 show photomi.crographs of the nickel deformed under various conditions. 11 The @@: work showed that in the first stage of creep.,deformation of grains on account of visible slip lines takes pla*ce by way of increasing displacement along slip planes and -age multiplication of these planesi in the second st multiplication of slip lines is the main factor. Deformation due to grain-boundary displacement occurs throughout the creep time and plays an increasing, part at higher temperatures and lower stresses. Certain boundaries can, depending on conditions7 either hinder-or stimulate deformation in grains. In addition to displace-, ment processes associated with slip lines visible under a microscope and with grain boundaries7. displacementloccurs through microscopically invisible slip lines and crack Card 3/3 formation. There are 11 figures) 1 table and'16 references: 5 of which are Soviet and 11 English. SUBMITTED: December 6, 1958 SOV/126- - -7-5-15/213 -AUTHORS: Rozonborg,.V1.M6 and Gradovav@L.N. TITLE: Changes in.the Struoturo.of Nickel and Solid Solutions of Titani um in'KickelDuring High Temperature Deformation (Izmeneniye struktury nikelya i,tvardykh rastTorOT titana nikele pri,vysokotemperaturnoy,deformataii), PERIODICAL: Fizika matallov i metallOlTedeniye, Vol 7, Xr 5, pp 722-731 (+ 1 plate) (USSR) ills-? ABSTRACT: Th thLs paper the natureof structural changes in nickel and solid.solution.8 of titanium in nickel during deformation at, high temperatures (1700 and .@90000) @Were studied.. @@.Tho composition*ofthe materials studied-is Ahown in.@Table 1.: In order to obtain uniform grain size of 092-Oo3.mm alloys wQro@initially annealed, at .1200OG Athe allo7 no, 134 for 1@hour, no.136 for 2 hours and no.137:for@3 Aoiira)o Theload was selooted so. that 'fracture should occur 'after a, few t9na of hours. That,deformation at which failure ooevorad after a few hours.is called rapid,.and that at which it o caurred., af ter. a f o.w tons,,of hours slow. ..The apoeimens were ..C.a 'Pulled in, a vacuum apparatus. . In Table 2 the,oonditions rd under,which the specimens wore pulled, and the a is santiia, W5 test resultal..are shown. The nickel spa eimen no.. 134-.1 Changes in the Structure of Nickel and Solid Solutions.of:Ti,tanium in, 'Nickel During High Temperature Deformation (the temperature forpulling was 7000C, -stress 5.2 k9/GM12) can be considered as an example which shows.thechangesin microstructursoccurring under the action of stresses which lead to rapid fracture. In Figs.l.and 2 photomicrographs taken at various stages.of,straining of the specimens are shown. FigQ3 shows the structure in the vicinity of_the, fracture. The specimens were also studied by X-rays. .. In all eases a few sharp points are so Lan in the X-ray photograph taken of anlundeformed specimen "(see Fig.4). In Fig.5 a series of X-ray @photographs is shown., taken of the same, point of the specimen at various stages of expansion. A similar change in microstructure,and X-ray picture obtained for.,,. specimen no.134-1 was al8o observed for specimen,no.154-2 (pulling temperature 700 Cp.stress, 3.5 kg/mm2) no.130.o2 (70000, 8 kg/mm2) and no.137-2 (7000C, 6 k /m;@). In the deformation under the influence of stress which leads to, fracture,after a few.tens of hours, the change of micro-_ Card structureldiffers from-the above ohanges,andis,notieeably 2/5 independe ,nt of the p.ulling,temper.atur es .used in the experimentso 7-5-15/25 Changes in the Structure of Nickel and Solid Solutions of Titanium 'in *Nickel During High Tomperature'Deformation. As an example, the deformation of spa 0imen, no 137-J5 (Ni + 603% Ti) at 90000 and stress of 2 kg/mm:4 is discussed. In Figs.6, 7@ and 8 the microstructure of a specimen aft'-er a 4%,deformation and fracture is.shown. Sub-grains appear not only in grains in whichintense slippage has occurred butlalso in those in'which only barely notiegabla signs of slip are evident. - When the,,sub-grains are noticeable in the slip "lines it can be seen that-there is al:definlto Pon- ma,tion between the direction of the slip:linea and thalsub- grains: the long side qf.the,latteris. perpendicular to the slip lines (see Figs.9@9 10 and 11). In. slow def ormatiou. the@appearance-of X-ray.photographs taken of the specimions also changes. in FIS.12:portions.of X-ray photographs taken of the 'specimen no.137-6 are shown. .@The changes in.atruoture described for 'specimen no.137-5,, tested at 90000 at a streas of 2 kg/=2@@are also observed for nickel. @ (90000 alid 2 ke/=2) Card and for nickel contain Lng 3.6% T-1 .(9000C, 2 kW=@;,' TOOuc, 3/5 g/mm2). In Fig. 13 a photomicrograph of the specimea no.135-7 4 SOV112-6 r? 1512 1@ 'Changes in the Structure of.Nickel and Solid Solutions of Titanium in Nickel During High Temperature Deformation (nickel 3.6% TI) after straining by 9.5% is shown,: This specimen was deformed at 7000C at a stress of.139,9 kg/mm2. At 700 and 9000C fracture of the specimen occlarred' along the grain boundaries at all stresses. As a result of the above investigations the authors have arrived at th4l, following conclusionag 1. In the deformation of nickel and solid solutions of, titanium in.nickel at high temperature4 formation of slip planes, fragmentation of grains.and displacement of grains- relative to each other can be Iobserved.,. 2. The fragmentation of grains in slow deformation at high, temperatures does not appear to be-the primary direet'breaking down process of the original grain. It is'associated with the preliminary non-uniform deformation of the latte--j. leading to lopalbending of,the crystal.. The observable fragmentation is a partioular.cass of polygonization occurring under the simultaneous action of temperature and deformation. _Ia the Card r.egion 700-9000C the possibility offragmentation is deter- mined essentially by the rate of@deformati.pn. 0-ov/126.. -- 7-5-15/2r; Changes in the Structure of Nicke.'1 and Solid Solutions of Titanium in Nickel During High Temperature Deformation 3. In the propeas of alloying nickel with titant-am, fra,#ent- ation of t-he grains oocura,at- lower deformation ratea than in pure nickel. 4. The fraeture of specimens tro3ted at 700.and qoCpc 4,31, Card inter cryst all i-ne in.nature.. 5/5 There are 14.figures, 2 tables and 15 references, of wh.. 4 are Soviet, ]LO English and I Swiss Institut metallo,redeniya fiziki metalloy.. T,51i'llChM (Institute of Metallurgy and Physics of Metals TsNIIOhM)'.@ SUBMITTED-, Jaauar7 20,.11958 sov/v6-8-2-24/26 ,m AUTHORS. Neymark, V.Ye. and Rozenberg, TITLE: Influence of Boron on Recrystallizati-on of Silicon iron PERIODICAL: Fizika metallov 1 metallovedeniye, 1959, Vol 8, Nr 211 PP 3,14 - 316 (USSR) ABSTRACT: The authors have studied the.influence of boron on the kinetics of therecrystallIzation of an'iron-silicon alloy with 39'a S!, 0-030% C, 0.2% Mn, 0.0190' P and 0, 0-003t 0-005 or 0.015'a B. X-ray methods were used,to investigate the cold-rolled (600' reduction) metal. /0 Figure 1 shows the relation,betweeii.tempera.ture and time for the start of recrystallization; the,dependence of the-data on boron concentration is.shown inFigure 2... The activation energy-rises. continuously with-increasing, boron concentration. For the.0.01% B alloy the. activation energy rises with decreasing temperature ofthe start of recrystallizatiow an effect similar to one observed by Rozenberg.with,E.Z. Kaminskly (Ref 2)@ and the authors suggest that this should be studied Cardl/2 further. 18M SOV/32 25- -?3/51 'A rJT'40R Rozenberg, V. M. TITLE: Evaluation of,the Contribution of Grain Boundaries in the General Deformation on Creeping (Otsenka vklada granits zeren v obshchuyu deformatsiyu pri polzuchesti) PEPIC)DICAL, Zavodskaya Laboratoriya, 1959, Vol 25, Nr 1, pp 53 57 (USSR) BSTP ACT Slow deformations taking place at high temperatures effect a grain displacement (with respect to one another). Temper- ature increase and decrease in stress usually bring about.. an increasein the relative deformation share with respect to,the grain boundaries-The formation of a lattice@j,by abrasion) permits a quantitative evaluation of the,deformation. in connection with the grain boundaries (Ref 1). On the basis@ of a schematic representation of a micro.structure with division lines (Fig 1) the deformation determinationis-ex- plained and the respectivecalculations are.:given.Jt.is- stated that the method has a few deficiencies;and,is,also C ard1/2' inaccurate., To ovoid these disadvantages the fact, is made Evaluation of the Contribution of Grain.Boundaries in SOIV/32J,25-1-23/51 General Deformation on Creeping use of that the grains do not displace only lengthwije. A schematic representation show!@-(Fig,4) that the.vactor of.. displacement can be divided into three components. Two of them are or the plane- of the saMDle surface.and the third one is perpendicular with recnect to the plane. The size of the vertical component can be measdred by an interferoneter according to Linnik (Ref 4), by the displacement of the interference waves at the' grain boundaries. Vertical displace- ments of the magnitude of 0.03-0-05 p can thus* be determined 'bration lattices. Calculation without the formation of calL examples, as well as a diagram of nickel creepin, at 8000 and of-a stress of 2 kg/mm2 (Fig 7) are given. A microscope of the 'e.TII-4 type was'used. There are 7 figures and 6 references, 3 of which are Soviet. ASSOCIATION: Tsen-tralfnyy nauchno-issledovatellskiy institut chernoy metallurgii (Central Scientific Research Institute of Ferrous, Metallurgy) Card 2/2 24(2) SOV/55-67-4-3/7 MIMS: 17adgorriyy,. E. M., Osiplyan,,Yu. A.,Perkas M. D., @Rozenber& T_(TLE: Thread-shaped Crystals With a Strength That Is Near Theoretical Strength (Nitevidnyye kristally s prochnostlyu, blizkoy k teoreticheskoy) PERIODICAL: Uspekhi fizicheskikh nauk, 1959, Vol 67, Kr 4, pp 625-662 (USSR) @'IBSTRACT: The present paper gives a survey of results obtained (especial- ly by papers published in.',Vesternperiodicals) concernin.- the properties and the growth of the so-called "whiskers", i.e. thread-shape crys-talsq whichg as regards order of magnitude, are 101 times as long as thick. The strength of these crystals surpasse's that of ordinary crystals of the same substance by 10 to 100 times their amount and.attains values that are near those calculated on the basis of the forcesof interatomic interaction. Special interest is further caused by investigations of electric resistance @especially -he ferro-_ at low temperatures), of the domain.structure-of t magnetic crystals, as@well as of photoelectric and optical Card 1/3 quantities. The present paper:presents a clear survey of what SOV@53-67-4-3/7 Thread-shaped Crystals With a Strength That Is Near T eoretical Strength ,ives d t 1 s has hitherto been achieved. Part I of-the-paper f, e ail (with numerous figures) concerning the, formation,orientation, and shape of-the whiskers; breeding by the regeneration of. metals from their salts, and breeding by means of condensation from vapors, and other methods are dacribed, as.also.the pro- duction of nonmetallic whiskers; a number of, phot ographs shows the shape and growth of copper- and tin-whiskers con- siderably enlarged (Up to.9000 times). Part II.contains a very vivid depcription of the growth of such..crystals as well as data concerning a large number of papers, which are given in a table covering two pages. Part III deals.with experiments and results concerning the mechanical properties of the whiskers; among other things, experimental data on the deformation of whiskers are compared with those of ordinary crystals; the tearing of these whiskers with as well as without previous plastic deformation is investigated and described in diagrams The creeping of metallic whiskers is described (also the:creep- ing resistance of whiskers is considerably greater than that, of ordinary crystals of the same material). Finally, the, influence exercised by temperature and by the.dimensions of Card 2/3 whiskers on "heir strength is described as also the influence "@O 0 68691 P.2 00 S/180/60/000/01/015/027 E193/E135 AUTHOR: Rozenberg., V.M. (Moscow) TITLE: the Relationship between Grain-Boundar3@Slipping and Deformation within the Grains during_2r @e PERIODICAL: Izvestiya Akademii nauk SSSR,Otdeleniye,tekhnicheskikh nauk, Metallurgiya i I@toplivo,1960,Nr 1@ pp:105-110 (USSR) ABSTRACT: It is a:well established fact that there is a close relationship between the grain-boundary slipping and. deformation of grains during creep .(Refs 1-3); the relationship is.linear and,holds for all stages.of creep, i.e. it is not dependent on the-rate of deformation The linear character of this relationship indicates clearly that,under certain given conditions, the process of creep in all its stages is determined by one mode of deformation. However, the question whether grain-boundary slipping.or the deformation within the grains,is paramount, remains still unanswered. MacLean and Farmer (Ref:l) who.had inves.tigated,creep of an age-hardening copper-berrylium-,:, alloy, and Martin et al (Ref 4) who had studied creep o Card brass, in erred fr 1/8 f om their, results that deformation within the grains is the governing process. In the fl 68691 8/180/60/000/01/015/027 E193/E13 5 The Relationship between.Grain-Boundary..Slipping and Deformation:. within the Grains during Creep opinion of the present author these conclusions. are not. justified and he considered.that more useful information could be obtained by studying creep at a certain constant temperature and.determining the part played by:both grain-boundaryslipping and grain deformation,on a specimen in which it would.be possible to vary the resis- tance of the grains.to deformation withou@,changingtheir size or chemical composition, It.was decided that the. 70Fe-3ONi alloy,would be most suitable,for this purpose. This alloy,constitutes a y solid Solution with a phase-, centered cubic crystal structure; on cooling below room temperature the alloy,undergoes a y to a transformationo similar to the austenite to martensite transformation.4 on subsequent heating above 5500C, an: a. toly trans.- formation, also martensitic in character, takes place. It has-been shown by several workers (Refs 5-7):that -the Card Y -4 ct --:-kY transformation.brings about@fragmentation of 2/ 8 the soli@ solution-grains,into.a large number-of small regions, which is accompanied. by distortion of the 68691 s/l8o/6o/ooo/ol/ol5/027 E193/19135 "The'Relationship between Grain-Boundary.Slipping and Peformation within the Grains during Creep crystal lattice at the boundaries of these regions: the metal in this condition is thuscharacterized.by the presence of alarge number of demarea'tion,planes (sub-, boundaries) in.the grains which-have.retained their original size, It has been shown, also. that the resis- tance to creep,(at 770. 0C) of specimens that have under- gone the @ y -@p m - > y transformation.is lower than that of untransformed alloy. Since raisin g:,the temperature of the preliminary annealing from 8,00 to 1100 OC:reduces the difference between the resistance to-creep of transformed and untransformed specimensl and s,ince.this effectis obviously associated with.the number of latticedefects due to the y --@:a -@,y transformation:being,reduced by a hi.gh-temperature,annealg itwas concluded that comparison of the total elongation in creep,with the elongation due to grain-boundary slipping,.measured on,t,ransformed and., Untransformed materialg would.make it possible to Card determine,the effect of the grain structure on the 3/8 behaviour of the grain boundaries in creep. :To make sure. 68691 S/180/60/000/01/015/027 9193/E135 The Relationship between Grain-Boundary Slipping and-Deformation. within the Grains during Creep that any of the observed,effects were.not caused by changes in the grain boundaties@themselvesl it was decided to carry out an additional series.of tests on single crystal specimens. The composition of the experimental alloy (prepared by vacuum melting technique) was: 30% Ni. 0.4% Ti,*0.15% Mn remainder.Fe. Single crystals were prepared by the pulling-up technique, the polycrystalline specimens by the normal, mechanical treatment methods., The single crystal.test pieces w .ere cut from one single crystal by means.of.spark-machiningi care being taken.to ensure@that the orientation of:both test pieces in respect.:to the axis of:load was the same: the dimension of the gauge length.of these test pieces was 2 x 1 x l0mm.: The dimensions,of the.gauge length of@the polycrystalline test pieces were 3*x 2 x:25 mm. All the, @test.pieces were~poli-qhedl.first.mechanicallyg,and then electrolyti6ally@in a: 95% acett anhydride plus 5% Card 'pieces were annealed hydrochloric acid.mixture.i All.test )+/8 at 1350 OP, for 2'hou'rs and cooled;in the furnace to r 68691 S/1806o/ooo/ol/o15/027 -E 193/E135 'The Relationship@between Grain-Boundary Slipping and Deformation within the'Grains during'Creep temperature (first type of heat treatment); part of the annealed test pieces was cooled in liquid nitro en Y -@-+ a t.ransformat-ion) and then.heated to 750 9C, held at,,that,'temperature,for 6 hours ( a -) y transformation) and cooled, in vacuumq t Ioroom temperature (second type of heat treatment). The creep tests were conducted in.: vacuum (lo-3 mm Hg) at 700 OC under,the stress of 2.5 kg/MM2; -periodically the tests:were interrupted to measure:the total deformation and (in the case of the polyarysta.lline sDecimens) the deformation due to the grain-boundary.movement. The creep curves forthe single crystal specimens,are reproduced in Fig 1, where curves a.. and b relate to specimens subjected to the preliminary heat treatment of the first and second types respectively. It will be seen that the annealed crystal deformed at a much slower.rate than the specimen which had Card undergone the y -) a -) y transformation. This indicates 5/8 that shortening of the time-to-rupture and increased rate of creep of austenite after martensitic transformation5 68691 S/180/60/000/01/015/027@@ E193/E135 The Relationship between Grain-Boundary.Slipping and Deformation within the Grains during Creep. The creep curves versus h) for,. 00 polycrystalline specimens, subjected to the preliminary 'heat treatment.of the first and second type, are reproduced in Fig 2 (curves A and b,,respectively); the -f (microns) versus V(h) curves for the same specimens are reproduced in Fig 3. Here again, the rate of.creep of the annealed specimens was slower than that of the specimens @that had undergone the :'y_--@ a -) y trans- formation.'@Since metallographic examination has failed to reveal any int6rgranular cracksq it.'follows that the deformation within the grain, 9cl@q is gi.ven by@ e the (microns).versus- relation- eck':= eo@_ r i@ eck ship.is illustrated by the graphs in Fig Thus, the foll wingfacts have been established., (1) the resistance 0 to creep of a solid solution that has undergone the -@.m --),y ttransformation is lower than that of the same @solution in the annealed condition;@,(2) since this card 7/8 decrease in the resistance to creep has been observed also in single crystals,,it can be assumed that it is mainly J J" 6 /6, - 0/00? /0@ '0 1 E-0 Si C- y-D T,1 AUTHORS: acs-enberg If i a;- ---------------------- TITLE: A St-ody of Grain-Bouardary Dis-plan-ement i n C' 1e e p PERIODICAL: F-izika metallov k1netallovedExiye !960 Vol Nr 1 p1p, 12L--132 (USSR) ABSTRACT; t, -n b -ou rvia r y The authors repo2rt their -in-vestigatioz) Df gral J- displacement due to an]-argemen' of ,-ome g--@a-ns at th e exDertse (.,,f' ol@hers d@-iring creep c!f' Figs 1 and 2 slhow tl-.@e also calliT b&-andary migration. The object -wras the elucidation cif tbe 01- u t e ect, in creeu and, the in,f'lulence of tixe boundary lo(-at-ion wizli '--espeCt to 'thle extension. axis and o f, t-1 h e mutual orientat-c the ex t ent -i-n of neighbouring grains or- (90.0(? of dispiac,@!mant- A-lumninium witb traces of Ylgl L Si and Cu was used e -@.,orking pa-.-t of a specimet e i ng I Lh n b x Y, 50 mm Annealinc- for 1 hour at LOO G%" gave a mean. grain diameter of 0.2 mm, unaffe,-ted by p@--olonged heating at 275-3-25 cc. Specimens @14-ith ground and, 0 1 e t-r-0 I Y t call surfaces were tested at_La pollstied Card of @O,Ir jCg/mg,2 COnstant load criving an initial st-ress 3@oecimens were removed from the machine at a carLai."'L. S/'11,2-_b/_-0/009/01/023/031 Bill/B191 A Study of Grain-Boundary D-J'splacament in Creep stage of Creer, and tho extent of mi.igration of 100 L -oundar es -vas n.-Leasured microscopically (differentiation fropi movement of one gra-in relative to anoth.er being e f fe c, t ed as shown schematically in Fig 3) C; r e e. p ourves th 300 and 325 OC. are shown in @ Fi obtained a t 2 corresponding @!alues of average displacement being gi,*,en _JZ! Fig 5, Fig 6 shows these values Dlotited asainSt deformatior, while Fig 7 givas relations between tha lo_@;-arit'hms of the ti'me to obtain di-S-placement's of 2@ and l.q and! the reel-procal of absolute temperat7ire. These re-iations.are linear-, ind,icating an a,-tivation_ energy of -1 K 188 kcal/moi, Use olf this value in the applropi-iate equation enables the curves of Fig rej-ating to three dif,"If'erent teirpe--ratures to represented by a sin,-:Ile clarve (Fig 8)@ kgreement- with- the dat-d of _Harpw@, Siiepard ancl Dorn (Ref D') is shown by Vine- reSults JL7r - FJ-g '91 Card t,1.7,c- -,.-alue of the. displal-emen-t.. a'Lcr-,-,a bo-un-da-'Y' 2/-) cvement s rela t v e to . ea -h @her) is :01 .-e- g7ain a L ve' a -@airis t ti angle bez a t@_,e spef:timer, axis and the ra r@@a_ 1 the.- boundary on the- CG and @a 6 i,54 b 3 S/126/6o/009/01/0'-3/103i E.11.1/El91 A Study cX Gra-in-,Roundary D.-ILsplacemarit- in Creep deformatica, o--:' a maxlurcam,ccc-drs cttl:'!-,,':,, OC, an(@ the 1--west, -,7alues a, t 0 ah,," 900 Fig: '-0. shows til@ abs@@nce of cori @---,atio-n be,",ween- this aii4s; and the cou,ndary T'01c- authors note tliat 1: t t j. 11 ur.certai-ri whothe-1, the toundarv, @Ls d.!sL'-,I.c3.c-.ed -f'roma gra In -le r to One nf p h; -1 -Gwer eiasti -ne--r,-y ur 'Tricc versa , -er t@ nt The n .T ;ornmand invest ation r, a poi dis-ol-a-cement eertairlY leads to a r e du. c t I o P- ir tntal 6h L, elastic energy of neighbouring ggrain-@ and 6@ @ 9 ma - er ia-, as a whr;le. and 1)1-asti@-!Lt y InCreases, The@.e al-a. 10 figures@ 2 tables and'. 9 referances, o --F whish Card are 3 Bngl-i-sh4 2, Fre-m'-th and, _7 frorin Act"a 3/3 Retal-lurgica. ASSOCIATION: Inst4tut- me tallof izikl; - TsNII(.Td,1 (Metall-physics (I,-SU1kute of Tsi-111cu") SUBMITTED: September 1959 8.0890 9073/E335 AUTHOR R enberg, V.M. "z n ITL T T E Slow Defo,,.tion %f Single CrXstals of/iron nickel Solid Solution at. Elevated Temperatures PERIODICAL- Fizika metall ov i metallovedeniye, 196o, V61 -9,- Nr 6, pp 922 926 (ussn) ABSTRACT: It was established earlier that depending on the test conditions phase work-hardening maIyIlead,to an increase or to a decrease in@.the strength. Specxmens,subjected to phase work-hardening as a, result.of loading,over-long 0 periods at elevated temperatures (700 C) have a lower. strength than specimens which were not subjected to phase work-hardening. :This difference is due to structural trans'formations,leading to a change in the grain properties, but it canalso-be due.to a change in the mechanism of deformation, whieh dependson the initial structure of the material. To verify the latter, it is adv1sable to investigate micro.sc.opi-cally the picture occuring in the case of.creep of solid-solution.specimens. To eliminate the influence of grain boundaries such investigations are Cardl/4 best carried:out on single crystals. In the experiments 80890 S/126/6o/o69/06/018/025' EOJVE3-@ 'Slow Deformation of Single- Crystals o, an I on-nickel Solid Solution at Elevated Temperatures described in this paper, an alloy containing 300% Ni Ti, 0-15%,Mn, rest Fe was used. An Alloy ofthis. composition represents asolid solution with a face-,. centred cubic lattice. On c,ooling the y-solid solution below room temperature a martensitle y4a transformation takes place. Heating of the transformed a-solld solution 0 above 500 C leads to a reverse, a-A-y transformation, which is also of the-martensitic type. The y-@a-37y trans- formation brings about afragmentation ofthe y-phase cilystals into numerous small areas,',disrupting the regular structure,at the.boundarles of these areas. Such chan Iges in the intracrystalline'structure lead to property changes. Two specimens, 2'x 1 x 10 mm', were.produc.ed by electro- eros IIon cutting of a single.-cr'ystal; orlqntation,lof both relative to.the tenstle-stress axis was equal. According 17t to analysis, sliding hese crystals should proceed in the plane , (111) in' the direction rl!01.. Both-specimens 0 were annealed at 1 ?50 @C.for two hours and cooled to room temperature in the furnace. Then one of the specim ns Card2/4 80890 s/i26/6o/oo9/o6/oi8/025 EOJ@OF-3R2 Slow Deformation of Single Crystals of an H7 ckel Solid Solution at Elevated Temperatures., was cooled in liquid nitrogenand subsequently heated 0 (in vacuum) to 750 C and held at that temperaturefor. six hours, thus transforming it into@the y-phase - The 0 specimens were tested at 700 @C with a tensile load,of 2 2.5 kg/mm Xn vacuum. Aftersome stretching, the specimen was relieved of the load, cooled and removedfrom the vacuum chamber..,The total elongation was measured and the surface was subjected.to metallographIc investigation. After fracture, the reliefoccurrIng.,at the surface of the specimens was studi ed, by means , of_ aq int@.erf erenc e microscope and,the.change in the orientation of.the crystals relative to the tensile-stress,axis was determined by X-ray diffraction. The obtained data lead to the conclusion that the difference in the behaviour of an annealed.specimen and a specimen which-was subjected to phas.e work-hardening is riot due to qualitative changes in the dbformation mechanism. Due to the presence of- Card3/4 lattice defectsin.the transformed-specimen the process of sliding will be more intensive in it thanin, anneale4_,.," specimens. 82641 S/126/60/010/02/011/02.0 E021/E335 AUTHORS: Gorelik,_S.S., Kallyanova, S.M. and,Rozenberg, V.M. @TITLE: Structural Changes in Aluminium With Slight Deformation and a Subsequent Annealing A PERIODICAL: Fizika metallov i metallovedeniye, 19609 Vol. 10, No. 2, pp. 251 - 261 TEXT: AV-000 aluminium containing traces.of magnesium, silicon and copper was used in the investigation. Deformation was produced on a 5-ton press at 21 mm/min. The number of slip .marks, the vertical component of displacement in the slip marks and the vertical component of displacement of grains relati--xe to:one another were measured.on an interference microscope by the method described in earlier work (Ref. 3). The mean grain size was also found. M�gration of the grain boundaries during re- 'crystallisation was found by a method.using pol@arised light (Ref. 5). The critical degre e of deformation using 1. an annealing temperature of 400 OC --ras.found:by..construct�ng a graph of grain size after annealing against de'gree,of deformation.-It was found, to be approximately 6-50"0. Fig. 2. shows the influence of small degrees.of deformation on the number of slip marks, the mean dis- Card 1/3 8261a s/126/60/010/02/011/020 E021/E335 : Structural Changes in Aluminium With Slight Deformation and a,, Subsequent Annealing placement in the slip marks, the mean displacement between the 'grains and the ratio of the.last two, in descending order. This shows that the basic.mech,anilsm.of,plastic flow, with or:without the criticaldegree of deformation, is the-same - slip in the grains accompanied by,displacement.of,the grains, relative to one: another. After deformati-on'less'than the critical value, siructural-changes during annealing occur, in the main, by polygonisation, with a small degree of,migration of boundaries of individual grains, stimulated@by the tendency to decr,ease the surface energy of a system (Figs- 3, 4).. Very occasionally, migration of the boundarj: occurs because of differences.in volume' energy of adjacent grains (Fig. 5). After-deformation greater than the critical amount, annealing-at 40b 'C is accompanied@by,., intensive growth of individual,crystallites and frontal,migration of boundaries which is stimulated by differences in.volume el astic energy of adjacent grains. The, rate and distance of migration of bounda ted ries is many times greater than.that stimulat by surface energy differencesu Intensive growth occurs afteran Card 2/3 82641 S/12,6/60/010/02/011/020 E021/E335 Structural Changes in Alumi,nium With Slight Deformation and a Subsequent Annealing @incubation period during -which,redistribution of theenergy inside the grains occurs (Figs. 6-8). After deformation many times greater than the critical amount, new grains arise between the original grains at places of maximum distortion. The orientation of the new grains differs sharply from those surrounding it. This occurs after an incubation period of 6 to 30 minutes. There are 8 figures, I table and 7 references, 6 Soviet and 1 English. ASSOCIATION.o TsNIIChM SUBMITTED: March 19, 1960 Card 3/3 ---------------- r- ACC NRt AimV SOURCE CODE B070 E070 58/65/000 0 AUTHOR: Rozenberg, V. M. TITLE: Fine slipping during creep SOURCE: Ref. zh. Fizika, Ab s. 12E537 REF SOURCE: Sb. tr. In-t metalloved. i fiz. metallov Tsentr. n.-i. ift-ta chernoy me- tallurgii, vYP- 36, 1964) 67-76 TOPIC TAGS: creep mechanism, crystal deformation, nickely physical diffusion ABSTRACT: The nature of the total-odeformation fraction (ix).whose tracks cannot.be. observed on the sample surfac%ly ordinary optical microscopy,was investigated in singleAnd polyerys'tals of Ni. Itis,established.that'rm varies with the temperature and stress in the same manner, as the part of the defo;m;fion due to the coarse slip' tracks. @x is 4 - 8 orders of magnitude larger than the rate of deformation estimated on the basis of the theory:of diffusion creep. These, and also some other data given, in the article, give grounds fora Iasuming that the.deforma It.ion whose tracks cannot be observed by metallographic ethodsp'is:the result of fine slip. V. Rozenberg. (Trans lation of abstract SM CODE: 20 Card Vj- '- v to . ACCESSION.,-., Frobler# la0allovedonlya--l- _s TOPIG-TAG&i al., gral 't t @4fidta r:rle t n: s rue,- ure. t on._due_*,---.t e opma, d'-f d f oonnebtbid-.Lw- dUh e:,@graiii-_,---b dair -e gr t is, png_, t 0 h!@,, 40j!@Ablo;-_ h tal d f6@' i'Vls@ e mat o --'iiot-@o_ilden this part as-'.Ex,-' '@.'T ilia crpop-:tradoes,@-.. d 'wi ofit lh r ur an ap r ntil'ditt..*.:fl-bih'. pe .6f, H7 lit -dtiird-, 6-@ thi k .-Andidit x s. ress ame-VAT.Iii 03 d on.-;t; e,-,, epe-n dibfide -0, -a.n-,i -A unste'ady- state,.,11. a-ge @'v-itli--@-iiAir -'@aga.. - - @A:@ s --Vith@-s(---constt wLta ViAu6s- e s-t:@tei - --t t"'- -- b `@4TLBA ows'." ha the@@,- c reated or a areal colt! And to Im iiiereased rate of ,@-d.ii confirmbd that deformatiori ii6ti ig.thpresult of melviTig temperii and.-,rAted,5( indiVi du an inc6n@ici6r- bl A--- P--.e-8i-. r i IAS.SO -0 T-, v@ SUBMITTEDC--@--*@ 0,0-@:@I@lf'-.t@:@'-- me all. tb a 0- me W776/62/000/024/002/007 EIII/E135 AUTIIURS Vinograd, M.I.? Rozenberg, VANI., and Shapiro, TITLE: Modern methods for phase analysis of steel and alloys SOURCE: Noscoly". TsentralInyy nauchno-is'sledovatel'skiy xnstitut chernoy metallurgii. Sbornik trudov ' no. 24, 190U. Novyye metody ispytaniy wetallov. 191-203. TEXT: Phase analysis is important. in developing new materials ,with special properties and in improving existing inaterials. The, authors outline the characteristics of four main groups of methods available: wetallographic, X-ray, chemical and electrochemical, physical. As examples of their application to the solution of, currently important problems the authors discuss tile following: i low strength of weld in tubes of type 1Y,1849S (lKhl6N98) steel; formation of sigma-phase in high-silicon steels and alloys, leading'. to loss of ductility; low plasticity in tensile tests on some heats of type X2-5 (Kh25) steel; excessIve inc-fusion content in type 0 )(18 Hi 9 T , (OhhlbNW), steel; estimation of.inclusion content in high-purity steels, e.g. type LUX15 (ShKhl5); failure of steel in hot mechanical deformation. Card 1/2 9754 Z S 26/62/ol4/oOl/010/018 E193/F,383 AUTHOR: Ro z enb erg.... V. N. TITLE: A study'of creep of nickel single crystals PERIODICAL: Fizlka metallov i metallovedeniye, v. 14, no..l, 1962, 114 120 T &,T: The main shortcoming of creep tests carried-out on polycrystalline specxmens as thatthe results'may be affected to various degrees by the presence of grain boundaries, and, consequently, may not be easy to interpret in terms of basic la-v-ts governin-,r theiphenomenon of creep. This prompted.the aut'nors to undertake the investigation described in the present paper.- To ensure the same orientation and degx7ee of purity in all 'the test pieces, they were cut from one" N! single crystal, p,repared by the conventional pulling-out technique. The creep tests were carried out in vacuum at 670 7700C under'a stress. 2 varying from 350 - 1 190 z/mm It was found that under the rate conditions employed the of steady creep is described by: Card l/ S/126/62/014/ool/olo/ol8 A study of creep of E193/E383 n E A exp (-67 OOO/RT) where, is the applied stress T temperature and n 4.4. This relationship was found to be in mood agreement with that based on Weertman's theory of clitab of dislocations, which means that diffusion-controlled rela_-,@at�on is the slowest of the processes determining the rate of steady creep. Using the Weertman equation and ex@eriment-al` data obtained in the course of the present investigation, the author derived formulae for the density of sources on-dis- locations 'M at. a constant temperatuzra: 9 M 2 x 10-76 2 cm-3 (5) and at a constant stress; exp(-38 000@T)@2 M 3 x 11035 (6) Tz @j C. .Card 2/ 4 S/126/62/014/001/010/018 A study of creep .... E193/E383 9 10 10 cm-3 for the t re- The values of M were 10 emperatu and stress ran-es studied. These values were in good agrreerjent with those calculated from data on the size of subgrains and: the degree of their misalignment, which confirmed the validity of the equation derived by the present author. Data on creep of. single crystals were correlated, in the final stage of., the Investi-ation,-with those obtained by Weertman and Shahinian (Tans. AIME, 19516, 206, 1223) on polycrys'talline specimens. The resul' ts - aE e reproduced in Fig. 6, where (Z Tep Q/kT)/LLI.. OC/min. cin is plotted against stress (dyne/cm here, -L and LI denote, respectively, the size of grains and subgrains., the tvlanZles and squares relating to creep@of polycrystalline specim6ns at temperatures indicated in the.insert,othe circles 670 - 730 C- 'It Will A rel'ating to creep of-single crystals at be seen that this relationship is similar for both'single- crystal and polycrystalline material, which indicates that the @presence of grain boundarles does not. give rise to the appearance 9f a sQparate.mechanism,..of deformation, making asubstantial contribution to the total deformation in creep. There are 6 figures., CArd 3/4 5/126/62/o14/001/010/018 A study of creep E193/E383 ASSOCIATION: Institut metallovedeniya i fiziki m.et:allo.v TsNIIClu%l . (Institute of Metal Science and Phys 3'. C S of Metals, TsNIIChM) FTJBIIITTED.- October 31, 19G1 (initially) 41 January 2 1962 (after revisio Fig. 6: 66000 raly-rica, -.4 &,goo*. n0,1UKPUCma1?,?b1 fo 81/000 j 0 crysft.13 W C OUCMt7" v10M 7 A /0 00 101 Card 4/4 S/126/62/013/006/009/018:' E021/E192 AUTHOR: ..Rozenberg V.M. TITLE: ..The nature of the part of deformation not detected in microstructure during the creep of nickel PERIODICAL: Fizika metallov i metallovedeniyo,'v.13, no.6, 1962, 894-900 TEXT: Experiments were carried out on nickel single crystals in order to show how much deformation was caused by visible slip. lines and how much deformation was not connected with observab] structural changes. Specimens (2 x 3 X 10 mm) were cut from a single crystal of nickel (99.95%) with the strain axis and the observed*surface disaligned by not more than 2-30'. -They were heated in vacuo at 1150 *C for 6 hours and electrolytically polished prior to test. Testing was carried out in vacuo at @670-770 *C with shearing stresses of:450 to 1190 8/mm 2. Occasionally the specimens.were unloaded, cooled, and after removing from the apparatus, the.elongation, number of@slip.lines on the &@Luge length.and height of slip lines were measured. The total elongation,was measured on a microscope andthe remaining Card 1/2 S/126/62/013/oo6/ooq/ol8 The nature of the part of deformation... E021/E192 measurements were carried out by an interference microscope. it was established that not-all the deformation is localised in the slip lines visible under the microscope. The deformation not visible, under the microscope (ex) is obtained bysubtracting the deformation due to observed slip lines (ese) from the total dePor-mation. In the investigated ranges of stress and 'temperature Ex is greater than Est. Both Ex and Est when plotted against time show a period of decreasing rate of creep and then a region of steady-state creep. For ex, es? and etotal, the rate of change of deformation with time is proportional to time to the power of 4 '4. The possible rate of creep due to diffusion mechanism cain be found, knowing the coefficient of diffusion fornickel. It is lower than' the exi3erimental values observed for ex by a factor of.,,8-9. Thus,, Ex is caused by crystallographic slip and...not by adiffusion, mechanism. There are*10 figures and 2 tabless ''ASSOCIATIONI Institut metallovedeniya @ fiziki metallov TsNIIChM (Institute of Science of tletals and Physies.oflltetaisl. TsN11ChM1) SUBMITTED: Septembez-.20, 1961 Card 2/2 S/126/62/,013/003/023/023 E021/El8o AUTHORS: Voloshina, L.A., and Rozenberg, V. M. TITLE: Influence-of the orientation of aluminium single J crystals on creep PERIODICAL: Fizika metallov i metallovedeniye, v-13, no-3, 1962,. 11711-476 TEXT: Single crystals of 99. 991/0' a luminium were, prepared by selective recrystallization. The gauge length of the specimen was 25 ameter 6 m:.q. -lith stress of 1164 g/mm2 nim and its di, a at 20@OC, a sample in which a large number of slip systems could, operate deformed much less than a sample in which the.probability of slip was, extremely small except in the main system. After ten ,hours the crystals deformed to 3.2 and 22..5@o* respectively. At 300 OC,,with a stress of 124 g/mm2, the behaviour was reversed... The results conf4.rmed that a large number of slip systems results in strengthening at low temperatures and an increase in deformation at higher temperatures. It had been shown (Ref.'6:,. D. MacLean, Grain Boundaries in Metals (Granitsy zeren v metallakh), DIetallurgizdat, 1960) that slip in grain boundary Card 1/2 influence of the orientation of ... S/126/62/013/003/023/02 @3 E02l/El8O regions of polycrystalline samples.,was complex. In these regions,.' in the case of a face-centred cubic lattice, up to.five families of slip can operate. Thus, at low temperatures, this would result in strengthening, and at higher temperatures in an increase in deformation. In this way, the effect of grain:size, on creep resistance could be explained without the idea of the "viscous behaviour$, of the boundary. There are 3 figures. Z' ASSOCIATION; Inst:Ltut metallofiziki TsNIICh1-1 (Institute of Physics of Metals, TsNIICh1-1) SUBMITTED; July 17, ig6l Cavd 2/2 S/032/62/028/002/024/037 B124/B101 AUEHOR: Rozenberg, V. M. TITLE; Evaluation of the values of elongation due to slide tracks PERIODICAL: Zavodskaya laboratoriya, v. 28, no. 2, 1962, 219-222 TEXT: Values of elongation due to the displacement of a material in one. track and the number of tracks per unit length of sample yield sufficient data for the calculation of elongation due to all tracks in p9lycrystal.- line materials and single crystals. The meandisplacement of.polycrystal- line materials in one track can be determined by measuring the.height of the tracks with the -.@interference microscope MHO-4 (MII-4). The, mean elongation p due to one track is p = 2.3 h V1 + where .6- total' . total is the elongation of the whole sample. Hence,the deformation due to all slide tracks is given by E 100%, where n is the number of slide st tracks per unit length of the initial sample. The schematic diagram shown Card 1 S/032/62/028/002/024/037 Evaluation of the values of elongation ... B124/B101 in Fig. 2 is used to evaluate with X,Y, and Z being a system of s orthogonal coordinates related to the sample. Theelongation is given by rhl+72 2 41 = - Ta -nA (5)9 and, for small values ofA, h 2 = Altanp. Hence, Al = hj/@(tan 2A - tan 2p M- With the optical microscope, however, the total height of the slide.track H=nh is measured, thecorresponding elongaa- @H@j@ + H2 2 2 2 tion beingAL=nAl, for whichtL and AL = Hl/(/tan A - tan P)j_ re- tan spectively. A is measured by an X-ray method. The angle P andits,change with,deformation were determined using stereographic:projection (Fig. 3). The values of AL due to one slide track were calculated from Eqs. (5).and (7) for slide on two faces of a nickel single crystal perpendicularto each other. The accuracy of H measurement is examined on a single crystal of nickel after 9.1%.deformation. H1 was found to be 0-113v which agrees fairly well with.results calculated from Eqs. (5) and (7). The method .,Card 2/4 S/032 '62/C,28/002/024/037, 1.1yaluation of the values of elGiigation ... 124/ loll i B. de3cribed can be used for coarse-grained materials, with thc, orientation of the individual grains being determined from X-ray data and etch patterns, rc'spectively. There are 5 figures, .1 table, and 5 references: I Sov Lit, t -language publicaLtion -reads and I non-Soviet. The.reference o -the English as follows: ft. A'Aaddin, E. flarrisong R. Gelinas, Act-a ;,i,et. , no. it, AS'o'OCI-ITION: Tsentrallnyy nauchno-issledovatel'skiy in,9titut chernoy metallurgii im. I. P. Bardina (central scientific Rezearch Inotit,,--Ae of Ferrous 'Metallurgy ihieni I. P. Sche@:iatlc diau@raln of step formation due to dizlacc-e'n't alon.- tile eli-u plane (p i.-i the displacement *vector; hV 'h2' --.-,.nd 1 are comr-onents of the displacement vector)- stereographic projection illustrating the methad. uscd to deter- 3. (V noi@m-zl mi--! the -angle :P. Legend: (A) plane of observation,,,. A to [he G~~.mple.ourface;.(C),-directidn of slip; (D) axi6.01 el oy')c7aw-' tioll-M slip plane. Gayd 3 /4'_' S-lLIRI,4'OV-AIYAYEV, Georgiy Aleksandrovich, prof., doktor tekhn. nauk5 GASTE%T, 2- 1 >92 S/126/61/012/001/014/020 E193/E48o AUTHORS: Voloshina, L.A. and Rozenberg, V.M. - ---------- TITLE., Investigation into the.creep of aluminium bi-crystals PERIODICAL: Fizika metallov i metallovedardye, 1961, Vol.12, No.1, pp.118-124 TEXT: The generally accepted view that the relative movement of crystals plays an important part in deformation during creep is based partly on the fact that grains stand up in relief on a preliminarily polished surface of a specimen tested'in creep. However, it can be postulated that this effect is caused'not by crystal boundary sliding but by the fact that the adjacent crystals are deformed in a different manner. The most convenient way of. checking this theory is to conduct creep tests-on specimens on which the. behaviour of any given grain boundary can be easily studied', bi-crystal specimens being most suitable for this purpos. e.. The effect of temperature, applied stress and degree of misorientation between grains on the rate of creep and degree of deformation of bi-crystals has been studied'by several workers. The object of the present investigation was to measure the elongation of various parts Card 1/7 25922 S/126/61/012/001/014/020 Investigation into the-creep ... E193/E48o of an aluminium bi-crystal test piece with.a@particular reference. to the grain-boundary regions and to study the micros Itructure of ,these regions. The bi-crystal specimens were. prepared from.. 'AVOOO) grade aluminium by the usual method of annealing the ASOOO olightly deformed metal. The creep test pieces (with the parallel ''portion measuring 25 x 5 x 3 mm) were cut from the blanks in such a way that the bi-crystal boundary was inclined at 45* to the test 'piece axis. The test pieces were annealed7at 6oo*C and electrolytically polished, after which a set of thin scratches was, ,inscribed with an aluminium alloy scriber on the polished surface All creep tests were carried out at 300% under an initial stress of 0.2 kg/mm2. Each test was periodically interrupted and, after co.oling.the test piece.toroom temperature, the total elongation of the gauge length, the elongation ofeach crystal and. the displacement of the scratches intersecting the grain-boundary were measured, Metallographic and X-ray diffraction analyses of the specimens were also carried out. Typical results are .reproduced in,Fig.1, where the relative elongation (%)'is plotted e (hours); this graph.was constructed for a test gainst tim . -piece 'whose stereographic projection is illustrated'in Fig.2, where Card 2/7 S/126/61/012/001/014/020 25922 Investigation into the creep E193/E480 continuous and broken lines relate to the first and second crystals respectively; the plane of projection corresponds to the specimen surface, the point in the centre of the projection representing normal to the specimen surface; the arrow indicates the direction of applied stress, The gauge length of the specimen was divided into,5 Parts in the way shown in the insert in Fig.l. Curve 1 in Fig.1 relates to the entire gauge length (i.e. to the part of the test piece bounded by points 1 to 6) while curves 2, 3,, 4, 5 and,6 relate to parts bounded by points 1-2. 2-3, 3-4, 4-5 and 5-6 respectively. It will be seen that the rate of creep varied' considerably from one part of the test piece to another. , It is particularly significant that the rate,of creep of the part containing the grain-boundary,@i.e. the part bounded by points 3- .4 (curve 4), was slower than that of the adjacent parts, bounded by points 2-3 and 4-5 (curves 3 and 5 respectively). Owing to the difference in the orientation, the rate of creep of any part belonging to the crystal bounded by points 1-3 was slower than the ,rate of creep of any part of the crystal bounded'.by points 4-6. The results of other measurements are best summarized by:referring to Fig,3. which shows diagrammatically the bi-crystal (a) before. Card 3.' 7 25922 S/126/61/012/Ool/ol4/,020 Investigation into the crrdep ... E193/E480 and (b) after deformation; the symbol I in Fig 38 denotes the grain-boundary displacement and x = 1 cos a is the component of resolved in the direction of applied stress. it will be seen from Fig.4, where x(ji) is plotted against the strain AE3-4G) of - the part of. the test piece bounded' by points 3-4, that x- is bylone order of magnitude smaller than &E3-4 and that.there is a, definite (linear) relationship between these two parameters. By, the process of elimination of the possible explanations of these facts, the present authors showed that the observed displacement of the grains is a result of the deformation of the grains in the, grain- boundary region and not -ice versa, This conclusion was supported by the results of the next series of measurements, whose results, are reproduced in Fig.6, where the vertical component of that is the height p of the step formed by the grain-boundary displacement, is plotted against time (hours-), blocks 1- 7 showing the magnitude of this component at various points along the grain- boundary, as markedon the insert in Fig.6. It will be seen that not only the height of the step varies along the grain-boundary but that there is a point.at.which the grain-boundary displacement throughout the duration of the creep test remains smaller than,at Card 4/7 25922 S/126/61/012/0,01/01.4rOO2O Investigation Into the creep ... E193/E480 any other.point; in addition@, the height.of the step at certain A p6ints*of the grain-bound'ary..decreases.with,.time lsee@point,3,after- 14hours, Point 5 afte'r 40 hours,,etc). These,results, correlated with the results of metallographic and.X-iray analysis, show conclusively that the formation of stepi;*at'the grain-boundary du.r.lng creep isnot caused by grain-boundary sliding (relative -0 ii vement of the adjacent grains) but is,mainly a result of.the- difference in the degree of deformation of the crystals.in,the rer,ions adjacent to the grain-boundary. Acknowledgments are made @to the student I.B.Finkellshteyn who participated in this work., There are 8 figures,..-l table and 8 references; 3 Sovket and' 5 non-Soviet. The four most.recent references to English langubLge-'@* publications read as follows:, , McLean D., Farmer M.H. J.Inst. Metals, 1956, October; Tung S.K., Maddin R., J.Metals, 1957;,9,:.' NO-7, Sec., 2, 905; Rhine+s F.N., Borud W.E., Kissel M.A.' 2__ Trans'. ASM, 1956, 48, gig; Intrater J., Machlin E.S. J.Inst-. M@tals, 1960, 88, No.7, 305. ASSOCIATION: Institut metallovederxiya i fiziki metallov TsNIIChM (Institute of I-fetal Science and Phys.ics of Metals SUBMITTED; Septer.-iber 9, @1960 TsNIIChM) Card 5/7 Wo. _001-A i N _09-4 0_6 80890 S/l26/6o/ooq/o6/ol8/O25: "E073/E335 AUTHOR: fRozenber"'_@ V.M. an Single Crystals of/lron-ni. el TITLE.: c@-S@j'o iv , D f ck Solid Solution at Elevated.Temperatures PERIODICAL. Fizika metallov i metallovedeniye7 1960, Vol -9-. Nr 6 pp 922 - 926 (ussn) ABSTRACT-. It was established earlier that depending on the test conditions phase work-hardening may lead toan Increase or to a decrease in the strength. Specimens subjected to phase work-hardeninglas Ia result of loading over long periods at elevated temperatures: (700 0C): have.a lower strength than specimexis which were not subjected to phase work-hardening. This difference is due to structural transformations leading to a change in thegrain properties but it can also,bedueto a, change in the mechanism of. deformation, whlch-.depends.on the initial structure of the material. To verify the latter, it is,advisable to investigate microscopically thel,picture..occ,uring in the case.of,creep.of solld-solution specimens.@ To eliminate the influence of:grain,,boundaries such investigations are Cardi/4 best carried.out ion single cry.stals. In the experiments 80890 .8/126/60/dog/o6/ol8/025 E013/E a 31&n-nickel Solid Solution Slow Deformation of Single Crystals o, n. at Elevated Temperatures described in this paper, an alloy containing,300/ Ni o.4-,,' Ti, 0.1550' Mn, rest Fe was, used. An @alloy of this composition represents a solid solution with a face- centred cubic lattice. On cooling the y-solid solution below room temperature a martensitic y,4a transformation takes place. Heating of the transformed a-soli,d solution. above 500 OC leads to a reverse, a-@,^( transformation, which is also of the martensitic type. The y-@a-7y trans- formation brings about a fragmentation of -the y-phase crystals into numerous,small areas., disrupting the regular structure at the boundaries of these areas. Such changes in the intracrystalline structure lead to property changes.- Two specimens, 2 x I x 10 mm, were produced by electro- erosion cutting of a single-crystal; orientation of both relative to the tensi-le-stress axis was equal. According 17these crystals should roce d 'n.,the to, analysis, sliding plane (111) in the @ direction . fi!01 Both specimens were annealed at. 1 350:'C for,two hours'and cooled to room temperature in the furnac e. Then one of the specimens Card2/4 80890 S/l26/6o/oo9/o6/ol8/025 EOJ@/U@@ckel.So Slow Deformation of Single Crystals of an n-. lid Solution- at Elevated Temperatures was cooled in liquid nitrogen and subsequently heated 0 (in vacuum) to 750 C and held at that temperature for six hours, thus transforming it into the y-phase. The 0 0 specimens were tested at 700 C with a tensile load of 2 2.5 kg/mm in vacuum. After some stretching, the specimen was relieved of the load, cooled and removedfrom the vacuum chamber. The total elongation was measured.and the surface was subjected to metallograpliic,.investigation.. After fracture, the relief occurring at the surface of the specimens was studied by means of aqint-erference microscope and the change in the orientation of the crystals relative to the tensile-stress axis was determined by X-ray diffraction. The obtained data lead.to the conclusion that the difference, in the behaviour of an. annealed specimen and a specimen which was.subiected to phase work-hardening is not due to qualitative changes in the dj@formation mechanism. Due to the presence,of, Card3/4 lattice defects in the transformed specimen the process of sliding will be more intensi-ve in it than in anneal specimens. 26561 S/126/61/012/002/011/019 ti EI021/E480 ,AUTHORS: Voloshina,-L.X., jozenberg, V._M_ and Finkel'shtevn, I.B..1. TITLE. The connection between boundary migration and' deformation in the boundary zones during the creep o If metals PERIODICAL: Fizika metallov i metallovedeniye, 1961, V.ol.12, No.2,'' pp.265 _268 TEXT:_ Experiments were carried out on bicrystals of aluminium. The boundary between the crystals was at 450 to the strain axis of the sample. A stress of 200 g/mm2 and a temperature of 3000C were used. Lines intersecting the grain boundary were drawn on. the electropolished surface before test. During the creep test, the relative displacement of the lines (in the plane of -observation) was measured by an-interference microscope. Fig.2 shows (in microns) the relative displacement of the grains (circles), and the migration of the boundary (triangles) against time (in hours). Fig.3 shows a microphotograph of the bicrystal taken on the interference microscope (x280) and Fig.4 the profile of the same- ,sample at the grain boundary. The data show that deformation in Card 1/4 s/i26/61/012/002/011/019 tlo5u The connection,between 61dary ... E021/E48o n the boundary regions is not stopped when migration of the boundary occurs., Thus, the alternate.relative,displacement of grains and migration of boundary, as proposed in Ref.l@(Chang H.C. Grant N.J. J.Metals, 1952, No.6, 619) to explain the cyclic character of the deformation-process in the boundary regions, is impossible. It is proposed that the cyclic nature of the process is caused by alternate hardening and softening, the softening in this case being connected with boundary migration. There are 4 figures and 11 references: 5 Soviet,and 6.non-Soviet... The four most recent references to English language publications read as follows: Chang H.C., Grant N.J. J.Metals, 1952, No.6, 61g; Rhines F.N., Bound W.E., Kissel M.A., Trans..ASM 1956 48, 919; Tung S.K.,,Maddin R. J.Metals, 1957, 9, N7, sec.2, 905; McLean D., Rev. met., 1956, 53, 139. -ASSOCIATION: Institut metallovedeniya i fiziki metallov,TsNIIChM (institute of Science of Metals and-Pbysics of Metals TsNIIChM) .SUBMITTED: 'December 19, 1966 Card 2/4 24478 S/126/61/011/006/004/011 E193/E483 AUTHOR: Rozenberg. V.M., ,TITLE: The effect of substructure on creep if nickel PERIODICAL: Fizika metallov i metallovedeniye, 1961, Vol.11, No.6, pp.899-909 TEXT: The effect of preliminary heat and mechanicaltreatment on the rate.of creep hasbeen quite extensively studied, Thus, for :1.11stance, D.McLean and A.S.L.Tate (Ref,l: Rev,met,, 1951, 48 No-10. 765) showed that polygonization of aluminium grains, attained by rolling and subsequent annealing, reduced,the@rate of, creIep of this nfetal at 2000C,by a factor of@200 to 500,0- Similar results were obtai-ned by other workers who, however,,did not investigate the elementary processes associated with th,e de f o riaa t ion in creep. The object of the present inve'stigation was to study tile effect of.substructure on both intra- and inter- crystalline deformationin creep of nickel, which had been selected for this.purpose because of the existenceof a large body of nsity of experimentil evidence.on creep of this.metal and on the de .sub-boundaries after plastic deformation followed by annealing. Card 1/9 S/126/61/011/006/004/011 The effect of substructure E193/E483 Nickel of 99.9% purity with traces of Al, Si, C Iu, Co and-Fe was used in the experiments. It was established in a series of .preliminary tests that if this malterial is plastically deformed, (at room temperature) to less than 6% deformation, no new.grains are formed on subsequent heating to 8000C. The flat creep ' test pieces used by the present author (with the gauge length measuring 40 X 3'x 2 mm) were polished and vacuum annealed at 11500C for 2-3 hours. after which the average grain-size was 0.165 mm. They were then plastically deformed (at room temperature) in tension at a rate of strain of approx 30 mm/min to attain elongation varying between 0 and 6%, annealed at 800% for 1 hour, and electrolytically polished. (In every case the average grain-size of the specimen remained equal to 0.165 mm.) The creep tests were all carried out in vacuum at 7000C and under a stress of 2.5 kglmm2, The tests were interrupted periodically and the test pieces, cooled to room temperature, were examined for total elongation, the number of grain boundaries and slip lines on.the gauge length and the magnitude of displacement of both the grain boundaries and the slip lines. The results are reproduced Card 2/ 9 24478 S/126/61/011/006/004/011 The effect of substructure F,193/E4183 graphically. (Unless otherwise stated, the number ascribed to curves,in the diagrams Iquoted,below denote the degree (6,%) of the preliminary plastic deformation of:the creep test Pieces.) The general,characte*r of creep of the various test pieces is'shown in Fig.1, where the elongation W is plotted against time (hours). It will be seen that with increasing degree of preliminary deformation the rate of creep decreased. Moreover, no transient stage of creep was observed in specimen s, preliminarily deformed ,to 6,3-@5%. The effect of such structure on that part of deformation whi.ch is revealed by slip lines is illustrated in Fig.2, where the number of slip lines per I cm of the gauge length (n,l/cm), is plotted against time. Here, in every case, n increased initially and then remained constant at a level which decreased with increasing 6. In Fig.3, the displacement at the-slip":lines (Fckili) is plotted against time (hours). The data reproduced,in Fig.2 and 3 were used to estimate the part played in creep of the, test pieces studied by coarse,slip; the -results are 'given in.Fig., where elongation due to coarse Slip (Eck,%) is plotted.against time, and in F'19-5 where the ratio Eck/EO45&&@ (ECS" denoting the- total deformation) is plotted against EOG&M'- The effect of Card 3/9 24478 S/126/61/011/006/004/01 The effect of substructure E193/E483 substructure, formed as a result of preliminary mechanical and thermal treatment, on the deformation of creep test pieces due to. relative movement of the grains is illustrated in Fig.6, where the the grains P (iL) is plotted against relative displacement of _ g time. In Fig.7, gap P (11) is plot ed against cogakN). elongation due to coarse slip a The above data on the tota nd elongation due to relative movement of the gralas were based on the results of measurements. The elongation due to fineslip could not be measured because the fine slip lines could not be observed under the microscope. However, the elongation due to fine slip E_x Is given by Ex CC)S-jq - (Eck + c.rp) and the effect of substructure on this parameter is illustrated in Fig.9, where ex is plotted against time (hours)i in Fi .10 Xx is plotted against E 9 @ 4 The results obtained were analysed in relation to the results obtained by other workers and the following conclusions were reached. (1) The creep properties of nickel are profoundly affected by preliminary plastic deformation (in tension),followed by annealing, since not only is the rate of creep reduced but the shape of the creep curves is altered. The higher the degree of.preliminar y Card 4/9 24478 S/126/61/011/006/004/011 .The effect of substructure E193/E'183 deformation, the smaller is the total elongation attained during creep in any given time. (2) This.effect of preliminary treatment (resulting in the development of substructure) is associated with slowing down of the Processes',.,of both coarse and fine slip in the interior of.the grains.- (3)@, The rela,t,j-ve displacement of the grains which can be observed I on the surface of..'@ creep test pieces is caused mainly by non-uniform deformation-of-- grains and not by the viscous flow alorg the gra!@-boun-a_ (4) The change of the behaviour of nick-el.during t@@L-ep-brought L about by preliminary mechanical and thermal t m-int is a -A t n .manifestation of the effectof tbe.resul_ta etwo@,,,. of dislocations obstructing' slip, which--c-o-nstitutes the-,main n There are 10 mecha ism of deformation in cre-ap f 44ires, 1 table and 25 references: 7 Soviet-bloc. and 11 non-Soviet-bl,@'@@. The' four most recent refe@@andes to English language publi 'tions read -T.R., Dunn C,G. Acta met. , 'E, 'N@o ,- as follows: Hibbard I 1956t .3, 3o6; McLean D., Fax-mer M.H., J'Ytns@l Dorn J.F. I AS_-M-11956, 255; Metals, -1956-57, 85, 41; Parcer, E.R., ASM, 1956, 227." -ASSOCIATION: Institut metallovedeniya i fiziki metallov .',j3NIIChk- Card 5/9 (Institute of Science and Physics of Metals lj;@@T f PHASE I Esmy, EXPLMT@M: SOV51IZ5 BhC,,rya L", 11, To.rly 41@ksandrovich, llo@tor of Pt,ysica =1 @Lth-t!cz; Yak6v Goln,Ainor; Vera Euxinskiy, Candl- do to of PhIsici and :%O.omatIco; @i@tor I'Ar;Dylovich Kardtn@k!y; Vludlilav. lu-Amirvvra Yritiknya, Candidate of phyeles and Kth-t-cZ, Ivauovlch Doctor of Technicel @-ience@; Y@rly Andreyevich Oaiplya,,; fl4rk D.-eyJovilch Porlkt5, Candidate -of Tortnical Sclcncoa; Vlhdimir Kolgoyvich Cv.didnt. of Tech- nical r-lenios; Na= lift4kovich S4niltr, Cart a o of M-I'MM-Ml@Scl-nata; Nadithd,' t Trofirovna Travina, Candidate or rkvaics and NAhvmatic5; and Lay X@arkovlch Uttvvkly, Candidate of To chnical Sciancna. Roate@,%vrrarll,a v fWcbeakom motallovadanii (Radiography in Physi-dul Y.tullogrelshy) Moscow, F@tallureizdnt, 1961. 368 @p. 5.200 coploa printed. Spon3zrl Gcaularstvennyy Sovot =M. Toontrallayy 1.0t1tut cho@rav@f metallurgii im. I P. Bardina. Inatitut Motallavd...Jy. i flzzil'i 14t.23W. Lt. (Titl. paeo)s Yu. A. BaCarystalcly; Ed. or publishing Pouzot lo.r. barite; Tech. Ed.r Y..B. Vynshtoyn. Card In, PURPOSF, This harAL-cok is Intendol for x-ray %vCtX!CIAA3 VOTkInZ in plant Inborn- tortes of the motallureical and machina-manufaeturing ir4ustry. 1. may also t. useful to technteal personnel in thq field of applied x-rmy diffraction on.lynts employed at sclantific,tachnical, and educational institutions. COMRAGM The hnndtook contAins best: lnrorvj,ti@on of the methods employed An metal- logr4phy. It consists of four pttti. Part 1 contains doscrIptiono or methods ror the study of polyerlst.13, Inoluiina tl,@ special features Of th* @Qrk' with oherp- focused tuboo and ionization countera, prepuration or speciza.", and choice of radiation ources, filters, comaraa, and g.0metry of the pict@e. Data on the photomaterIng of x-ray Pictures and on the application or olactrem, diff'raction techniques to matal sclenzo are alai presented. , Part 11 contains a detailed d6rcription of stresses and drrormatlen3 in crystals of metal, as well as of now motho,is for mfavarine tho alto or j@s!ns and areas of coherent amatterine. The material Alsa contains data art mothMn for stuAyIne the recry4tallization of estate for dettminine textures. Part III is devoted to x-ray phase analysis to be cerr%PJ out with the aid of tables included In the appendix. ?art IV deals with x-ray atu.1- Joe of sttol ttAt has been variously treated by thor=al and No raroonalities are mentioned. There are 222 rafer6nm4st 199 Skvlet, 55 EnCli.h, .26 Gorman# and 2 French. Card W7 e g n -Z J 19 A g v A Z A 810 .10 w 96' .0 Ile 3 -0 u sa 5 5 So 4 0 g .1 AN 3 t; C-Is q U-JA 0.0 Z Id a- 0 cr rd W. 8 00 0 0 g 0 q 0.0 a 0 0 a 0 0 0 0 & a 0 0 o o a a 0 a . I p 6 1 1 1 # -, - 1) J3 14 .7 2# It w It Pit hilt affaL I, Is uptill a , A r L _t I L, L a f & A so ., Of %0 I, ot rjl@ct of " - i N. Rolt,lll@ljf. J. it genic d1hydroxide. 04 0 s 14,) 14. 1 lit I 4(;A, 00 IIV dolt. tio' 1.,1.41 'I't-w -00 it, Ilils-til'. .'It "Wte uli-. it that tit. jtmi@tlv Of tit,- Nl_ i, 00 I,v @Al Imum1wit. Sunv -.111V lilt. ill., lm@ ill miltkilml twists 111)11t@ Ow 00 0 "llih. ilw %.lit, &, Ip't illitite thi@ ill, (.*%I, tkv., 0 11 . Ift. 11"flitt,'l 1w .,411 1'" 111.1twil, : M G Nt...'v zoo 0 .00 00 go t 09 rm I I 0 p 0 00 0 0 0 0, 0 0 0- - a 0 0;0 1 0 0 0 0 0 0 0 0 0.0 0 0 0 - -O 0 "i S/080/63/036/001/026/026 D2o4/D307 AUTHORS- Motsarev, G.V. and Rozenberg, V.R. TITLE: On the problem of preparing bis(trichloro- methyl)dichlorosilane PERIODICikL: Zhurnal prikladnoy khimii, v- 36, no. 1963, 231 232 ,TEXT: The present work is communication IV in the serie3 of papers dealing with the halogenation of al-4phatic silanes nnd siloxanes. Ris(trichloromethyl)dichlorosilane M was prepared in h14% yield by passing gaseous C12 into a solution of 210 F C13C.CH 3S'C'2 in 20 9 C12, at 600C, over 66 hrs, at 6.1 g/hr, using azo-bis-iso-butyronitrile as an initiator. CC13 Sici3and CCI 3' CHC12siC12 formed.as by-products. Compound I coiild also be prepared in CC14 (not less than 5 mol %) using uv irradiation as the initiator, at 20 - 250C, over 66 hrs, at 6.1 Cl I. CC130sici 2/hr. The latter method yielded. 60 % of 3 Card 112 ACCESSION NR- AP4032495 910080/64/037/00'410747/0749 AUTHOR: Motsare'v, G. V. ;_11oze0e R. TITLE: Thermal decomposition of t richlo romethyltrichlo roe ilane. Communica- tion, LK SOURCE: Zhurnal prikladnoy khimii, v. 37, no. 4, 1964, 747-749 TOPIC TAGS: trichloromethyltrichlor,osilane, thermal stability, thermal de- composition, mono chlo romethylt richlo ro s ilane, dichloromethyltrichlorosilane ABSTRACT: The thermal stability of the chloro derivatives of methyltrichloro-', silane was studied. The mono- and dichloromethyltrichlorosilane are complete- ly stable on boiling,.even in the presence of anhydrous FeC13. Trichlorome..thy14 @richlorosilane readily decomposes on boiling (160-175C) to form a 90% yield O'@ S' 14 and 1311/46 yield or C2C14- 0 ic rig. art. has- 2 equations. ASSOCIATION: None SUBMITTED: 18Mar63 DATE ACQ: llMay.64 ENCL: 00 SUB CODE: GC @NO REF SOV: 002 OTHER: 001 Card I/ I ACCESSIM NR: AP40325W S/0080/64/037/004/09W/0922 AUTHOR: Motoarev G. V.; Rozeaberg, V. R. TITIE; Initiating the chlorination reaction of methylehlorosilanes with a cetyleyelohexylaulfonyl peroxide. Communication VT_T in the series. SOURCE:- Zhurnal prikladnoy kbimji., V. 37j, no. 4j. 1964., 920-922 TOPIC TAGS: chlorination, mathylchlorosl2ane,, acetylcyclohexyloulfonyl peroxide,"-'-'' initiator, chlorination iniator, chloromethyltrichlorosilane, polychloro- M@thyltrichlorosilane, dimetbyidichlorosilane, polychlorodimethyldichlorosi3Aiae azobisisobutymnitrile, ultraviolet light ABSTRACT: The chlorination of methyltrichlorosilane and dimethyldichlorosilwe iniated by acetylcyc1ohex;yUulfonyl peroxide was investigated. Reaction proceeds li very low at 24-30C. At 50C with a@CH SiCl *C12 molar ratio of 1:0-3 t i here it I I , low 3- complete conversion of the chlorine naifig I weight part of the,monochlor- to 1.15 parts of polychloromethylsilanes. Reducing the peroxide ratio to 1:0.2 reduces the polychloro derivatiYes to'0.84 parts.,.The initiating action@ipf this peroxide exeeeda that of azobisisobutyronitrile-and approaches that of ultra- Card 1/2 ..... .