Institute of Metals, 1935, Vol

Home , Hiduminium

691, 6 9 2 & 6 9 3 U l to be printed in the half-yearly volume : Journal of the Institute of Metals, 1935, Vol. LVI.

Vol. 2. P a r t 1.

The Monthly Journal of the INSTITUTE OF METALS

and METALLURGICAL ABSTRACTS

JANUARY, 1935

'AZ' BRAND C.C.R. BRAND CADMIUM SELENIUM 99-95% GUARANTEED

THE BRITISH METAL CORPORATION LT-D

PRINCES HOUSE, 93 GRESHAM STREET,

■Phon«: METROPOLITAN 0111. LONDON, E.C.2 BIRMINGHAM SWANSEA

'Phone: CENTRAL 6 4 4 1 . ’Pbon.: SWANSEA 3166.

4 Alda’ Welding Rods and Fluxes The B.O.C. stock only those materials which they have proved by test to be most suitable for customers’ requirements. Large or small quantities of any material can be supplied. • Alda ’ rods are in convenient lengths, packeted in different diameters from -jV' to The range comprises rods for almost all steels and irons used; rods for copper and alloys, brasses and bronzes, and aluminium and alloys. 4 Alda ’ fluxes are obtainable in £, and i lb. tins. For every type of 'A lda’ rod supplied there is a suitable flux stocked. Special fluxes for brazing, tinning, and hard and soft soldering are also obtainable. Catalogue Section 4 listing the range of ‘Alda’ welding materials (with

H e a d O f f ic e s : V ic t o r ia St a t io n H o u s e , L o n d o n , S .W .i There are 41 B.O.C. factories in Great Britain and Northern Ireland and 23 factories in Australia? India and S. Africa. ______719/G The illustration shows one of the windows in the Taj Mahal Hotel, Bombay, where all glazing sections including the astragals are of aluminium alloy. The BRITISH ALUMINIUM Co.. Ltd.

HEAD OFFICE: ADELAIDE HOUSE, KING WILLIAM ST., LONDON, E.C 4. TELEPHONE: MANSION HOUSE 5561 & 807^ (5 LINES) TELEGRAMS: CRYOLITE, BILGATE, LONDON LO N D O N W A R EH O U S E: 25-29, Pancras Rd., N .W .I. BIRMINGHAM 4: 2, Lawley St. MANCHESTER : Deansgate House, 274, Deansgate. LEEDS 3 : 66, Kirkstall Rd. NEWCASTLE-ON-TYNE I : Milburn House. G LA SG O W C.2: 113, W est Regent St. A iii For the economical production of tubes with absolutely uniform wall thickness we supply TUBE EXTRUSION PRESSES

of vertical or horizontal type, also the required power water plants consisting of pressure pumps and accumulators, the latter of the electrically controlled, com pressed air loaded type without any pistons or floats, for which the H YD RAULIK Co. owns patents in all countries of the world.

British Agents: Aabacas Engineering Co., Ltd., 10 Canning Place, Liverpool, I iv o r k t & a l t

In addition to tubes, we manufacture rolled non-ferrous metals and extruded bar and sections. W e are the sole makers of DURALUMIN in various forms. (Regd. Trade Mark)

JAMES B O O T H £ » COMPANY (1915) LIMITED ArgyleSt. YJorks.Nedtells. BIRMINGHAM,7. telephone:EAST/22KP.EEJ. ‘Jet&jramr-'Booth,Birmingham. Codes: Cieber's 5letter. A.R C bth.Edit ion. Bentley's Second Phrase. ELECTRIC

FOR THE ■■■■ .

View o f a G.E.C. Furnace installation fo r 5-TRłP IN IKON. the Grünewald process at the works of STEEL,COPPER the Hall Street Metal Rolling Co., Ltd., PHOSPHOR- BRONZE. ETC Birmingham. THE GRÜNEWALD PROCESS ENSURES A PERFECT ANNEAL —ATHO ROUGHLY UNIFORM AN NEAL—AND A PERFECT BRIGHT BRIGHT FINISH. ANNEAUNG- 8UT M O R E ,A G.E.C. experts are at your service THOROUGHLY to advise on the most suitable UNIFORM ANNEAL equipment for any melting or heat FOR ALL PURPOSES treatment operation. Manufacturers— THE GENERAL ELECTRIC CO., LTD. Head Office: Magnet House, Kingsway, London, W .C.2. Branches throughout Great Britain and in all principal markets of the world.

vi ELECTRIC FURNACES FOR TUBE ANNEALING with uniform heat radiation from all sides by means of embedded heating-elements in self-supporting thin-walled oval muffles. Multiple zone automatic temperature control. Heat-insulation practically exclusively by powder, dispensing with brick-work. Motor-driven door. WELDED CHARGING FRAME of light construction, 8— 12% dead weight. Note the double charging troughs, facilitating re-charging.

Tube-Annealing Furnaces in a large Metal W orks; output 10— 15 cwt. per hour. Charge weight up to I ton. Useful dimensions : 23 feet x 28” X 12” and 23 feet X 35" X 12". Heat-loss when maintaining at 600'C.: 7 KW .; when at 480 C .: 4-8 KW.

Tube Furnaces were recently delivered to four leading firms ; each of these four firms has now given a repeat order. J. L. BROOKLAND WATERLOO HOUSE, WATERLOO STREET, BIRMINGHAM, 2 'Phone: Midland 4242. ’Grams: 44Brookland, Birmingham.’'

vii VITREOSIL FOR THE LABORATORY

The properties of VITREOSIL are ideal for the electri cally heated muffles necessary for controlled heat treatment tests. VITREOSIL basins, crucibles and beakers are heat and acid proof and are economical in use.

THE THERMAL SYNDICATE LTD. VITREOSIL WORKS, WALLSEND-ON-TYNE London Depot: Thermal House, Old Pye Street, S.W.I STERLING METALS!"

ELe kntc. no. 5tr ON

A typical Elektron Rear SAND & DIE Axle Worm Casing CASTINGS

Telegram s: Stermet Phone Coventry Telephones : 8035 (3 lines) for Melting Iron, Special Steels, and Non-Ferrous Metals, Is now built in England by

who have acquired its sole manu facturing rights for Great Britain, the Dominions, and certain other countries .... a Furnace of World wide reputation in Modern Foundry and Ingot Making Practice now

BIRLEC THE NAME SYNONYMOUS WITH PROCKSSS AND QUAlITt

BIRMINGHAM ELECTRIC FURNACES LIMITED B irlec w orks. Tyburn road. Erdington. Birmingham

ix a l l o y s , s l o u g -h Gram s:

DEGAS Y A L U M IN IU M WITH DEGASER No Eliminates Gases and Pinholes Produces Finer Grain, Cleaner Surfaces Increases Tensile and Elongation WE ARE MANUFACTURERS of a W ide Range of SCIENTIFIC PRODUCTS For the treatment of Non-Ferrous & Ferrous 'W M E T A L S A To Produce Clean, Sound Castings f A Ig f o f e . DEGASING, DEOXIDISING, CLEANSING LONG ACRE, FOUNDRY NECHELLS, SERVICES, LTD. BIRMINGHAM, 7 X ¡SÉS IEME■NS-SCHUCKERT

ELECTRIC PRE-HEATING FURNACES FOR w B K Ę Ü i l t t i SlSÄSS EXTRUSION

120 Kw. PRE-HEATING FURNACE FOR ALUMINIUM ALLOY BILLETS

Electric furnaces are especially suitable for pre-heating non-ferrous metals prior to extruding. Uniform and yet rapid heating, automatic temperature control, absence of corrosive gases and consequent elimination of waste are features of the electric system.

Advt.of Siemens-Schuckert (Gt. Britain Ltd.) 30/34 New Bridge St. London E.C.4. Phone Central 8461 XI A Guide to Your Journals !

C O N T E N T S

list of lTS PUBLICATIONS WITH m jth o r 'N ° ex

This Contents List, issued in a stiff cover, with an Author Index, will facili tate reference to the papers published in the Journal since the Institute’s foundation. It should be found most valuable pending the issue of the 10-year Index. Copies are obtainable from the Secretary. Price 2s. 7d. each, post iree. From our wide range we are able to supply refractory mat erials of high quality suitable for most industrial purposes. In our various works, which are modern in design and equipment, care is taken in every stage of manufacture to ensure that our products are maintained at a uniformly high standard.

I'or fuller particulars, ask for our Pamphlet, No. 1.

BLUEBELL Jokit (¿.Stein ¿Co. Lid. J tHISTU j ; NETT L E. f’- 'IC S tc o ^BOW NyBRIDGE SCOTLAND

xiii INDEX TO ADVERTISERS

JANUARY, 1935

PAflK Aluminium Plant & Vessel Co., Ltd Hydraulik G.M.B.H. iv Amsler & Co., Alfred

. . . might have been the first . . . no difference, every foot is the same . . . correctly tempered, exact width, splen did surface, clean edges, right gauge . . . very much cheaper in the long run to specif/ LUSTER Rolled Metal. • LUSTER ROLLED COPPER, BRASS DIPPING, GILDING AND CARTRIDGE METAL, PHOSPHOR BRONZE, ETC. J. F. RATCLIFF (METALS) LTD. NEW SUMMER STREET BIRMINGHAM

xiv BaW l D

Two Wild-Barfield Forced Air Circulation Furnaces for the Heat -Treatment of Aluminium Alloy Castings

ELECTRIC FURNACES for ALL FORMS OF HEAT-TREATMENT

A wide range of standard and special designs is available. Full particulars will be furnished upon application.

WILD-BARFIELD ELECTRIC FURNACES LTD. - ELECFURN WORKS - NORTH ROAD - HOLLOWAY - LONDON - N.7 JOHNSON MATTHEY & COMPANY LIMITED

where the resistance to tarnish, wearing and electrical properties are superior to the ordinary Base Metal Contacts.

Let us quote fo r your next requirements. Our assist- mice and service are at your disposal.

78 HATTON GARDEN LONDON, E.C.i.

XV i T . l o o ¡Z 5 ~

THE E d i t o r i a l O f f i c e -. A d v e r t i s i n g 36 Victoria Street, D epartment: London, S.W.l. Monthly Journal of the T. G. Scott & Sos, Ltd.

T e l e p h o n e : 63 Ludgate Hill, London, E.C.4. OF TELEPHONE : Editor : C ity 4211 (2 lines). SHAW Assistant Editor: ETALS S. 0. GUILLAN

V olum e JANUARY, 1935 Part 1

CONTENTS PAGE New Year Message from the P re s id e n t ...... 1 Institute News and Announcements ...... 2 Personal N o t e s ...... 5 Local Sections N e w s ...... G Papers to be Read at the March Meeting : 691. "Type Metal Alloys." By Frances D. Weaver, B.Sc. (Mrs. Harold Heywood) ...... 9 692. "Some Properties of Tin Containing Small Amounts of Aluminium, Manganese, or Bismuth.” By Professor D. Hanson, D.Sc., and E. J. Sandford, B.Sc...... 43 693. "Alloys of Magnesium. Part II.-—The Mechanical Properties of Some Wrought Magnesium Alloys." By W. E. Prytherch, M.Sc. . ¡59 Meetings of Other S o c i e t i e s ...... 77 Author Index to "Metallurgical Abstracts” ...... 78

METALLURGICAL ABSTRACTS I. Properties of Metals . . . . 1 II.Properties of Alloys .... 7 III. Structure (Metallography ; Macrography ; Crystal Struc- ture) ...... 14 IV. Corrosion ...... 17 V.Protection (other than Electrodeposition) . 20 VI. Electrodeposition ..... 21 VII. Electrometallurgy and Electrochemistry (other than Eloctro- deposition) ...... 22 VIII. Refining. . 23 IX. Analysis ...... 23 X.Laboratory Apparatus, Instruments, &c. . 26 XI. Physical and Mechanical Testing, Inspection and Radiology 27 x n . Temperature Measurement and Control 29 XIII. Foundry Practice and Appliances 29 XIV. Secondary Metals : Scrap, Residues, &c. . — XV. Furnaces and Fuels ...... 30 XVI. Refractories and Furnace Materials . 31 XVII. Heat-Treatment ..... 31 XVIII. Working ...... 31 XIX. Cloaning and Finishing .... 32 XX. Joining ...... 32 XXI. Industrial Uses and Applications 34 XXII. Miscellaneous ..... XXIII. Bibliography ...... 37 XXIV. Book Reviews ...... 40

The monthly issue of Metallurgical Abstracts may bo cut up for card indexes, as members will receive early in 1936 the year’s abstracts in bound form. xvii Fry's supply 7 0 /i

Printing Metals used in this country=

RY’S supply two-thirds of the M E T A L Printing Metals used in this Fcountry. FOUNDRIES During tl' 1: st q uarter oi a century grc advances have been made in the manufacture and con trol of the alloys used in the Print ing Industry. Fry’s metallurgists have played large part ¡'-1 efi'c i ii. M erton Abbey these reforms.

The alloy ; mploycd are of man, L o e d o e , s .W .1 9 grades and asses, the printer ca now rely u . r getting a suitabl Branches at Ma n c h e s t e r metal for 'cry purpose of li BRISTOL erait (whet err it be for a soli tan GLASGOW Linotype o ,.\Ionotype machine i. DUBLIN a remote r, untry printing office or a great in Hation for a leading daily newspper).

The full (’sources of the Chemi cal and Re. arch Laboratories of Fry’s are n the service of every printer. Tc production depart Manufacturiers ment is thi finest of its kind Europe. Evry batch [,[ m etal is subject to vtiemical ancl physical o f inspection and ;s guarantee«." Fry’s expe technical service is P rie tí eg M etals continuous.

W b ite A etifrictio e M eta i!j xviii A NEW YEAR MESSAGE

I n the year just ended the new method of publishing all papers in the Monthly Journal has been in full operation and its effects can now be gauged. Under this scheme, papers selected for oral discussion appear in the Monthly Journal well in advance of the meeting at which they are presented. Thus opportunity is given for careful reading and even for exper imental work designed to verify or to develop the conclusions o f a paper. The many opinions I have heard have been consistently favour able to the new plan and agree in largely ascribing to it the high level reached in the keen, informed, and unflagging discussions at the March meeting in London and the Autumn meeting in Manchester. This is one important matter in which the Institute, moving with the times, has further improved its service to members— and (perhaps unfor tunately from one bqint of view) not only to members but also to increasing numbers o f those ip \e metallurgical industries who are able to make f till use o f the Instiiute'j publications at no cost to themselves. The develop ment o f libraries, public and private, and the combination or closer associa tion of hitherto separate industrial concerns, can undoubtedly have the effect of making on set of Journals, fo r which the Institute receives one membership Subscription, accessible to larger numbers o f readers. Those who tab idvantage of this attractive possibility of receiving something fo r notlii f , especially those who can afford to pay their very moderate share of. t e cost, may well consider whether they may not in , time be killing the r, ose which lays the golden eggs. Happily, the Insti tute is still full of lif , and the mechanism which provides the golden eggs, in the form of sett life papers, continues to function vigorously, but1 the process of under'ceding, if it continue, is bound to have its natural consequence. This issue o f the Journal will first reach those who are, as members, paying subscriptions and thus providing their part of the cost of running the Institute and i publications. Perhaps they may find, in this brief message, an argutrn '< to use with their friends who have not ye t joined us. The Institute's 'resent need is a substantial increase of membership i to repair the gaps aused by the recent industrial depression and by the difficulties which have developed in international relationships. Those who bring in new 'members will render the best possible service to our Institute.

H, President. A 1 INSTITUTE NEWS AND ANNOUNCEMENTS ANNUAL GENERAI. MEETING, LONDON MARCH 6-7, 1035.

'['h e Twenty-Seventh Annual General Mooting will bo hold in the Hall of the Institution of Mechanical Engineers, Storey’s Gate, Westminster, S.W.l (by kind permission of the Council of the Institution), on March 0 and 7. The following is a time-table of the Meeting : Wednesday, March G. 10 a.m .- General Meeting. The Report of tho Council and the Report of 12.30 p.m. tho Honorary Treasurer will bo presented. The results of the election of tho Council for the year 1935-1036 and of now members will be declared. Papers will be read and discussed. 12.45 p.m. Members who so desire will lunch together at St. Ermin’s Restaurant (table d'hôte meal, 2s. 6d.). i Papers will bo read and discussed. 7 p.m. for Annual Dinner and Dance at the Trocadero Restaurant, Piccadilly 7.15 p.m. Circus, W.l. Thursday, March 7. ' 1 ‘>*30 "p m Papers will be read and discussed. 12.45 p.m. Members who so desire will lunch together at St. Ermin’s Restaurant {table d’hôte meal, 2s. 6d.). 2.15 p.m.- A visit will bo made to the Battersea Power Station. (Private 4.30 p.m. bus loavos St. Ermin’s Hotel at 2.15 p.m. prompt, return faro Is.) Certain of tho papers that have been accepted for presentation at tho Annual General Meeting on March 6 and 7 have already been published in the Monthly Journal (beginning with the Octobor issue). In tho next issue of this Journal other papers will bo given, together with a complete list of tho papers to be presented. There will be circulated to members in due course a form which may bo used to apply for tickets for tho Annual Dinner and Dance. Annual Dinner and Dance. The Annual Dinner will bo followed by a Danco which will bo arrangod as in recent years. The price of tickets for tho Dinner and Dance (exclusivo of wines) is 15«. Application for tickets may bo made jiow by members who desiro to reserve seats or tables for the Dinner. Discussion on Cold-Pressing and Drawing. In the evening before the Annual General Meeting thero will be held in the rooms of the Royal Geographical Society, Kensington Gore, a Joint Meeting to discuss tho “ Problems of Cold-Pressing and Drawing.” The meeting is being organized by the Institution of Automobile Engineers in co-operation with the Institute of Metals and the following other societies : Iron and Steel Institute, Chemical Engineering Group, Institute of Fuel, Institute of Marine Engineers, Royal Aeronautical Society, Institution of Mechanical Engineers, Institution of Petroleum Technologists, Society of Engineers, Junior Institution of Engineers, Institution of Production Engineers, and the North-East Coast Institution of Engineers and Shipbuilders. Tho discussion of tho subject will bo opened by two papers to bo read, respectively, by Dr. H. J. Gough, F.R.S. (Superinten dent of the Engineering Department of the National Physical Laboratory) and by Dr. Cecil H. Deseh, F.R.S. (Superintendent of the Metallurgical Depart ment of tho National Physical Laboratory'). Dr. Gough will deal with the mechanical problems involved and Dr. Desch with tli>. metallurgical aspect. 2 Institute News and Announcements

The meeting will bogin at 7 p.m., and it is hoped that there will bo a large attendance of members of tho institute of Metals. As the meeting is being held on the evening previous to the Annual General Meeting it may bo possible for many members from the provinces conveniently to arrange to attend both meetings. Members who expect to attend the meeting on March 5 can be supplied with copies of tho papers then to bo read on application to Mr. Shaw Scott. Election of Council. As only sufficient, nominations to fill tho vacancies announced a t the last General Meeting have been made, no ballot will be necessary, and tho following members, who have been nominated for election on the Council, will be declared duly elected at the Annual General Meeting . President. H a r o l d M o o r e , C.B.E., D.Sc., Ph.D. Vice-Presidents. W. R . B a r c l a y , O .B .E . C e c i l H . D e s c h , D .S c ., Ph.D., F.R.S. Members of Council. Lieutenant-General S i r R o n a l d C h a h i.e s , K.C.B.. C.M.G., D.S.O. Engineer Vico-Admiral S i r R o b e r t D i x o n , K.C.B., D . E n g . R . G e n d e r s , M.B.E., D.Met. A. H . M u n d e y . Tho Hon. R . M. P r e s t o n , D.S.O. H. B. W e e k s . Tour to Germany. Plans are now being completed for tho previously announced Educational Tour for junior members of tho Institute, college students, and membors of staffs of educational and research institutions. Tho tom- will bogin on Saturday, April C, the party being due back in London on Monday, April 15. During the visit—which will bo to tho Rhineland of Germany—several im portant works of metallurgical interest, ferrous and non-ferrous, will be inspected. Opportunities will also bo afforded to meet German students, as well as to sco something of tho beauties of tho Rhino. Already over twenty applications have been received from persons who desire to participate in the tour; those who liavo not yot sont in their names should communicate promptly with the Secretary. The cost of the tour is not expected to exceed £12. May Lecture. Professor W. L. Bragg, F.R.S., who is to deliver tho Twenty-Fifth Annual May Lecture on Wednesday, May 8, has chosen as his subject “ Atomic Ar rangements in Metals and Alloys.” It is of interest to roeall that Professor Bragg’s father, Sir William Bragg, F.R.S., delivered the Sixth Annual May Lecture in 1916 on “ X-Rays and Crystal Structure, with Special Reference to Certain Metals.” Autumn Meeting. The Institute’s Twenty-Seventh Annual Autumn Meeting will bo held in Neweastle-on-Tyne from Monday, September 9 until Thursday, September 12, thanks to a warm invitation received by the Council from members and friends of tho Institute on Tyneside. This will be the first occasion on which tho In stitute has revisited Newcastle since 1911. Tho members will receive, a wel come from tho Lord Mayor and tho subsequent proceedings will bo on lines closely resembling those of recent Autumn Meetings. Hostel accommodation will bo available for seventy persons. Members are invited to offer Papers for tho Autumn Meeting. Tho publica tion of accepted papers will begin in the April issue of the Monthly Journal. 3 Institute News and Announcements

Manuscripts should bo submitted in duplicate. Drawings for reproduction must be submitted in India ink on Bristol board, tracing or good drawing paper, and should bo drawn approximately twice the size that they will bo when reproduced: very largo drawings should bo avoided, if possiblo. A ll lettering must be in pencil to enable the Institute’s draughtsman to add this to conform to tho style adoptod throughout tho Journal. Photomicrographs must bo trimmed to one of tho following sizes, preferably the former: 2 in. x 2£ in.; 3 in. x 2} in.; 3 in. x 4 in. Magnifications should bo given in all cases. A brief synopsis should bo given at the head of each paper submitted, and tho MS. should bo accompanied by a declaration of originality, tho form for which may bo obtained from tho Editor, who should be advised as soon as possiblo by intending authors of papers of tho subject of the papers and when their MSS. may be expected. Membership. Forms of application for membership received by tho Secretary by noon on February 14 will bo considered by the Council on that day, and candidates whoso names are then approved by the Council will be able to take part in the Annual General Meeting on March 0. Candidates elected from now until May will havo tho privilege of membership for the extended period ending J une 30, 1936, instead of for the usual twelve months. Members are invited to acquaint friends who are qualified for membership of this concession. Forms of application for membership aro to bo found in each bound volumo of tho Journal and in tho now illustrated folder “ The Institute of Metals : Its Aims and Objects,” copies of which will bo gladly supplied by tho Secretary to any member or forwarded to a prospective member. Tho following wcro elected to membership on December 13, 1934 : As Members. Adams, William Oaklands Melbourne, Australia. B e a t t y , Alfred Chester .... London. B e s c i i o r m a n , William Charles New York City, U.S.A. B r e c k p o t , Professor Raymond Blanden, Belgium. B u r s t a l l , Aubrey Frederick, M.Sc., Ph.D. London. C l a u s s e n , Gerard Eden, M.E., S.M. Sheffield. G u i l l a u m e , Dr. Charles Edouard . Sevres, Franco. v a n H e t e r e n , Willem Jacob Utroolit, Holland. J o n e s , Reginald, M.Sc. . . Slough. K o s t e r , Professor Werner, Dr.-phil. Stuttgart, Germany. M a l k o v s k y , Jaroslav A., Dr.-Ing., D.Sc.. Prague, Czechoslovakia. M a t t e r , Jean . Paris, Franco. M i l l e r , Richard Franklin, A.B., Sc.D. . Now Haven, Conn., U.S.A. N i c h o l l s , H erbert .... Hounslow. P u l l e y n , Erie . Hayes. ScmMMEL, Alfred, Dr.-Ing. Finow/Mark, Germany. S e d d o n , Harold . . • Bradford. S i l j e h o l m , Gosta, Dr.-phil. . Stockholm, Sweden. S p r i n g , Kenneth Michael Uplands, Swansea. T i n d a l e , H arold . Sydney, Australia. W a r r i n e r , Lewis Legrand Milwaukee, Wis., U.S.A. W e i b k e , Friedrich, Dr.-Ing. Hannover, Germany. As Student Members. B l a u , Osmar Julius ..... Sydney, Australia. F i n c h , George Arthur . ■ London. H a l l , William Leslio . St. Albans. H u r s t , Harry Mackinder .... Sheffield. M a r k s , Edward Hubert, L.D.S., B.D.Sc. Melbourne, Australia. S u o w e l l , Geoffroy Dugard .... Birmingham. T i i o m a s , Gordon James, B.Sc. Wembley Park. 4 Personal Notes

The Late Dr. Rosenhain’s Papers. The Appendix to the Autumn Lecture on “ The Work of Walter Rosen- huin ” which appears in the Docombor issue of tho Journal (Volume LV, No. 2, 1934) constitutes a bibliography of tho published works of tho Institute’s Past- Presidont. Thanks to the kindness of tho oxecutors of tho late Dr. Rosenhain, copies of most of theso papers have boon presented to the Institute. The Council has docidod to offer—for their personal use—ono copy of each paper to thoso members who apply to the Secretary and oncloso postage and packing fees at tho rate of Id. per papor. As tho stocks are small—ranging from 2 to 50 copies —and are expected to bo rapidly exhausted, those members who apply promptly aro most likely to secure tho papers they request.

Leverhulme Research Fellowships, 1935. Application is invited for (i) Fellowships or (ii) grants in aid of research. Tho Fellowships or grants aro intended for senior workers who aro prevented from carrying ont resoarch work by routino duties or pressure of other work. The duration of tho grants will not normally extend over more than two years and the amount will depend on tho nature of the research anil t ho circum stances of tho applicant. Any subject which may add to human knowledge may bo proposod for a Fellowship, but proforonco is given to subjects in which other provision for resoarch is inadequate. Forms of application may bo obtained from tho Secretary, Dr. L. Haden Guest, Lovorhulmo Research Fellowships, Union House, St. Martin’s-lo-Grand, London, E.C.l. Telephone: National G701. Applications m ust bo received on or before M ardi 1. 1935. Awards will bo announced in July and tho Fellowships or grants will date from Soptombor 1,

PERSONAL NOTES

The Editor requests that his M r. N o r m a n E, D b n s e m , M.Sc. attention be directed- to items of interest Tech., has been appointed a Research to members that might be included under Follow in tho Department of Glass the “ Personal Notes ” heading. All Technology, Sheffield University. contributions for the February issue of the Monthly Journal should reach him D r . E u g e n V a d e r s has been ap not later than January 25. pointed Director of tho Forsehungs- M r. G. W. A u s t in -, M.Sc., was zentralstelle der Vereinigten Doutscho gazetted O.B.E. in the New Year’s Motallworko A.-G., Frankfurt-am- Honours List. Main.

M r. J . G. B e r r y ^ B .S c ., is in Jugo D r. W. R. W'h i t n e y has boon slavia assaying for a gold prospecting awarded tho 1934 Edison Medal of company. tho American Institute of Electrical P r o f e s s o r W i l l i a m C a m p b e l l Engineers “ for his contributions to has been elected Honorary Chairman electrical science, his pioneer inven of tho American Society for Testing tions, and his inspiring leadership in Materials’ Committee on Non-Fer research.” rous Metals and Alloys after serving for over 25 years as that committee’s M r. T h o m a s A. W r i g h t , Secretary Chairman. Ill-health has compolled of Lucius Pitkin, Inc. and President Professor Campbell to give up tho of the Buffalo Testing Laboratories, direction of a committee of which ho was recently elected President of tho has been Chairman over since its Association of Consulting Chemists formation in 1909. and Chemical Engineers (U.S.A.). LOCAL SECTIONS NEWS

SYNOPSES OF PAPERS TO BE READ IN FEBRUARY. London Section. The Protection o! Metals "by Coatings. By W. H. 0. Vernon, D.Sc., Ph.D. (Feb. 14.) Joint Meeting with Eloctrodepositors’ Technical Society. The author •«•ill discuss: (i) Natural protective coatings—i.e. “ natural varnishes” (invisiblo oxide films—stainless steels) and “ natural paints” (e.g. greon patina on copper), (ii) Various types of artificial coatings. Those regarded as outside the discussion include paints, varnishes, enamels, bitumin ous compositions, cements, rubber, &c. (iii) Methods of reinforcing or reproducing natural protcctivo coatings, e.g. oxide films by chromato dips and by anodic oxidation, green patina by chemical and electrochemical (anodic) treatment, (iv) Chemically-produced coatings other than those under (iii), e.g. phosphate and nitride coatings on iron and steel, alum-dichromate and selenium coatings on magnesium, (v) Metal coatings. General principles. Methods of production, with special reference to recent developments : (a) by rolling (cf. old “ Sheffield plate ” ), e.g. coatings on stool and on aluminium alloys; (6) by hot-dipping; (c) by cementation (Shorardizing, Calorizing, &c.); (d) by spraying (“ metallization ”); (c) by oloctrodoposition. (vi) Conclusion. Various factors—adhesion, thickness, structure, &c., towards the elucidation of which discussion may bo usefully directed.

North-East Coast Section. Melting Fuinaces lor Non-Ferrous Metals. By C. L. Cassidy. (Feb. 12.) The development of the crucible and crucible furnace will bo traced briefly from their early Egyptian origin, and the modem cruciblo discussed in rolation to service obtained under different conditions and mode of failure. The design of various types of furnaces, with reference to efficiency and running costs; relative advantages of coke, oil, gas, and electricity as heat media; furnace atmosphere and its effect on metal quality and metal losses; and special fur naces doveloped for latest foundry requirements, such as alloy cast iron, and die- castmg, will bo dealt with. Scottish Section. Recent Developments in Electric Furnaces ior Non-Ferrous Metals. By A. G. Robiette, B.Sc. (Fell. 11.) The progress and development achieved during the past five years by the application of electric furnaces to tho melting and heating of non-ferrous metals ■nail be described. After a brief review of the status of the electric furnace in tho non-ferrous industries, both arc and induction fumaco molting practice aro described. In the largo brass rolling mill the latter typo of plant is now the standard molting unit. Tho arc furnace has been little applied in this country, especially tho rocking type, which is widely used in America for tho melting of such diverso alloys as brasses, bronzes, nickel alloys, bearing motals, &e. For tho molting of aluminium and its alloys both in the foundry and for • wrought shapes, resistance furnaces can offer innumerable advantages, and their application with particular reference to porosity and other melting problems is discussed. Resistance furnaces for tho annealing of non-ferrous motals aro rapidly gaining favour for operations in which accurate temperature control is essential. The intermediate and final annealing of aluminium, copper, brass, and nickel- brass in the form of shoet, strip, wire, and other wrought shapes is being exten sively conducted in furnaces heated by nickel-chromium elements. The use of forcod-air circulation for hoat-treatinont processes requiring a 6 Local Sections News low tem perature (below 600° C.) is becoming standard practico, and lias m ateri ally incroaaed the efficiency, uniformity of temperature distribution, and rate of heating obtainable with this typo of equipment. One of the most outstanding developments has been the application of con trolled atmospheres to the heat-treatm ent of non-forrous m etals; and with fow reservations it is now possible to eliminate pickling from the processing of non- forrous motals. Tho plant and artificial atmosphere conditions for this work are reviewed in detail as well as economic considerations. Sheffield Section. The Manufacture of Pewter. By F. Orme, M.Met. (Feb. 8.) After briefly tracing tho history of tho pewteror’s craft, from Roman times until its decline in the eighteenth century, the paper refers to tho recent revival in tho demand for “ Pewter ware.” Tho various compositions and standards of both English and foreign pewter, the control formerly exercised by the Wor shipful Company of Pewterers, and tho old methods of manufacture are reviewed. Tho constitution and metallurgy of modern English pewter, a tin-antimony- copper alloy, are described, with special reference to the operations of melting, casting, and the various processes of mechanical treatment, rolling, spinning, hammering, &c. The effect of changcs in composition, and tho influence of impurities on tho properties of tho metal, will rccoivo consideration. Solder ing and the numerous “ surfacing ” operations, which present problems to tho manufacturer, as well as troubles arising from blistered sheet and cracking during spinning, receivo attention. Tho oxtont to which tho properties of tho alloy may bo modified by suitablo hoat-troatment is explained and will be demonstrated. Swansea Section. Design of Rolling-Mills for Cold-Rolling of Metals, Both Ferrous and Non- Ferrous. By C. E. Davies. (Feb. 12.) Tho paper deals with tho design of cold-rolling mills and tho principles on which modern designs are based, especially with reference to mills employed in the cold-rolling of strip and shoots of steel and tho commonest non-ferrous motals and alloys. Tho subject is considered under tho following headings : (1) Tho rolling mill as a machine; its essential features as represented by the simple two-high mill. (2) The theoretical principles involved in the cold- working of metals by rolling; comparison with alternative and analogous procosses. Illustration by diagrams of tho action of rolling. Force and speed relations, rate of deformation, calculation of rolling loads and power, &c. Roll material and deformation under load, deflection and camber. Roll bearings and bearing friction. (3) Chief types of two-high mills employod for various classes of strip and shoot rolling. (4) Rolling speeds, with refer ence to modern British, American, and Continental practice. (5) Detail mill design for modem requirements considering strains resulting in heavily loaded components. Accessory equipment and tho special importance of coiling gear in strip-rolling. Automatic coiling. (6) Tho backed-up roll mill. Advantages of the small diameter roll in the four-high and “ cluster” designs. Local Sections Papers. Tho following papers, read sinco October last before Local Sections of the Institute, have been deposited in tho archives of tho Institute and can bo loaned to interested members on payment of postal charges.

D e v e k e u x , W. C. “ Methods of Manufacture and Their Influence on Design. Aluminium Alloys, Wrought and Cast Parts.”* (Scottish Section, November 12.) * Appears also in slightly condensed form in Met. Ind. (Lond.), 1934, 45, 489-493, 513-516. 7 Local Sections News

G o u g h , H. J. “ Fatigue in Metals.” (Birmingham Section, November 20.) G u a y , Kenneth, “ Somo Manufacturing Faults and Other Defects in Extruded Load Products.” (Sheffield Section, October 19.) J o h n s t o n , f t . G . “ Directionality in Somo Annealed Alloys.” (Birmingham Section, November 29.) L a m b e r t , W . “ Manganeso-Bronze.” (London Section, December 5 .) P e a r s o n , C. E . “ The Flow of Metals in tho Extrusion Process.” (Sheffield Section, November 9.) S c o t t , A. W . “ Rhodium Plating and Its Applications.” (Sheffield Section, December 14.) S m i t d e i x s , C. J . Chairman’s Address. (London Section, October 11.) S p i t t l e , A. “ Improvements in Surface Condenser Tubes.” (Scottish Section, December 10.)

8 PAPER No. 691. This paper Is copyright. It may bo reprinted, wholly or in part. In the Press (with duo acknowledgment) after being presented at a meeting of the Institute to be held on March 0-7,1035, in the Hall of the Institution of Mechanical Engineers, Storey s £ Q t Gate, Westminster, London, S.W.l. The Institute as a body is not responsible for the U «7l, statements or opinions expressed in this paper, on which written discussion may be sent to the Secretary not later than April 1, 1935. This paper will not be reissued in the form of a separate “ Advance Copy,” a method of publication which has been discontinued. TYPE METAL ALLOYS.*

By FRANCKS D. WEAVER,t B.Sc. (Mb s . HAROLD HEYWOOD), M em bbb.

Synopsis. The microstructuro and properties of lead-base antimony-tin-lead “ type metal ” alloys have been investigated by means of thermal analysis and microexamination. The liquidus surfaco for alloys containing up to 24 per cent, antimony and 14 per cent, tin has been constructed. The general lines of the constitutional diagram put forward by Iwasé and Aolu have been confirmed. The existence of a true ternary eutectic in the lead-base corner is confirmed, but with the composition ant imony 12, tin 4, and lead 84 per cent., solidifying at 239° C. The ternary peritectic invariant point of Loobo and contemporary workers is shown to bo the eutectic point of a pseudo-binary system of lead and the compound SbSn. A method of etching has been devised which distinguishes between the a and (3 antimony-tin phases, whether present as primary crystals or as eutectic constituents. The microstructures obtained with different rates of cooling through the solidification temperatures, including those of industrially cast types, have been examined and compared. Hardness tests have been carried out on the alloys.

I ntroduction. The researches described in this paper were undertaken to investigate the microstructure and physical properties of the antimony-tin-lead alloys used in the printing industry and usually designated as “ type metals.” These alloys cover a wide range of composition in the lead- base corner of the ternary diagram. Type metal alloys are used for a variety of purposes, consisting essentially in the casting of exact replicas of brass or papier mâché designs, the replicas or castings being used subsequently by the printer to reproduce ink impressions on paper. The essential properties of type metal alloys are : fluidity when liquid, ensuring easy flow into the mould; absence of contraction on solidification, giving exact reproduction of fine designs ; low melting point and small range of temperature during solidification ; and good resistance to the wearing action of the printing process. The lead-base antimony-tin-lead * Manuscript received November 10, 1934. f Formerly Assistant Metallurgist, Monotype Corporation, Ltd., Rcdhill.

Note to Abstractors and Other Readers.—This paper will be published, in permanent form, in the Journal of the Institute of Metals, Vol. LVI, 1935. Reference should accordingly bo as follows: ./. Inst. Metals, 1935, 56 (Advance copv). 9 Weaver : Type Metal Alloys alloys give a good combination of these properties, the hardness and melting point being increased with higher antimony and tin contents. Type metal alloys include monotype, linotype, intertype, stereo type, &c. ; the first three being used in mechanical type-casting machines. The standard monotype machine casts single type varying from 5 p t. X 21- pt. (0-07 x 0-035 in.) to 72 pt. X 72 pt. (lxl in.), both of height 0-975 in., and the super caster from 5 p t. X 2i pt. (0-07 x 0-035 in.) to 72 pt. X 90 pt. (1 X 1-25 in.). The linotype and intertype machines cast slugs of from 5 pt. X 4 erns (0-07 X 0-67 in.) to 48 p t. x 36 eras (0-67 x 6 in.) (maximum) dimensions. The metal cylinders cast from papier mâché impressions of assembled type and blocks used in newspaper printing are made of stereotype metal. A detailed description of the respective methods of casting has been given by Mundey, Bissett, and Cartland.1 Typical compositions of the various alloys, found by experience to give the best results, are given in Table I. T able I. Antimony, Tin, Per Cent. Per Cent. Monotype metal : Ordinary .... 14-19 7-10 Monotype display . . 16-24 8-12 Linotype metal . . 10-13 2-5 Intertype metal . . . 11-14 3-5 Stereotype metal . . . 13-20 2-10 When this work was commenced knowledge of the exact structure of these alloys was incomplete and inadequate for practical purposes. The results of no thorough and reliable investigation had been pub lished for many years and the only constitutional diagrams available did not fully explain the phenomena and difficulties observed in industrial practice. Alloys used on slug-casting machines such as the linotype are the softest of all type metals, as low melting point and small range of temperature during solidification take precedence over hardness. The size of the linotype slug permits the use of a weaker alloy but demands easy and rapid solidification. It was well known 1 that linotype alloys solidify at an approximate constant temperature of 238°~239° C., but the generally accepted diagram due to Loebe gave no ternary eutectic near that composition. The only invariant point in the lead- base corner of Loebe’s diagram was a ternary peritectic of the com position antimony 10, tin 10, and lead 80 per cent., solidifying at 245° C. Linotype alloys have a eutectic structure, whereas monotype and stereotype alloys consist of hard cuboid crystals embedded in a eutectic matrix, the relative proportion of crystals to eutectic increas- 10 Weaver : Type Metal Alloys ing with antimony and tin content. The methods of etching avail able, however, could not distinguish between the two possible antimony-tin solid solutions nor determine the exact constituents of the eutectic matrix. Consequently, although the rapid cooling of the monotype casting system permitted the use of harder alloys with higher melting points, the compositions of the alloys used were only ascertained by practical experience and were not deduced from theoretical reasoning. Although it was assumed that equilibrium conditions could not be attained without very slow cooling, the relationship between the microstructures produced by slow and rapid cooling through the solidification range was uncertain. It was well known that all these alloys tend to segregate and that the crystal size increases with slower solidification, but comparison of the identity of the micro-constituents had not been made nor had the variations in microstructure produced by the different rates of cooling occurring in industrial practice been investigated. In a series of investigations, the results of which have been recently published in Japan,14 the whole of the antimony-tin-lead system has, however, been re-examined with special reference to printing metal alloys. A new constitutional diagram has been produced in which the lead-base corner differs considerably from that of earlier investigations, and which offers a more satisfactory explanation of the problems described. The investigations described by the present author confirm the modifications introduced by the Japanese workers but are made in greater detail over the area involved; they also pro vide additional information on the nature of the microstructures obtained with the different rates of cooling used in these investigations and in industrial methods of casting in printing machines.

H is t o r ic a l . Before studying a ternary system the three binary systems involved should be considered. The systems lead-tin, antimony-lead, and antimony-tin have been re-investigated recently and are well estab lished. A brief resume is given below. L ead-Tin.—Eosenhain and Tucker,2 in a very complete thermal and microscopical investigation, found the liquidus to consist of two curves intersecting at a eutectic point having the composition lead 37-07, tin 62-93 per cent, and solidifying at 180° C. They found that lead at this temperature retains up to 16 per cent, of tin, but that this solid solubility decreases to 7 per cent, at room temperature. An arrest point at 149° C. on the lead side of the eutectic composition 11 Weaver : Type Metal Alloys

was attributed to ail allotropie transformation. Other investigators later confirmed the results of Bosenhain and Tucker with slight modi fications of the solid solubilities, but failed to support the existence of horizontal lines below the solidus. The most recent diagram due to Stockdale 3 is shown in Fig. 1. The eutectic composition is lead 38-14, tin 61-86 per cent., and the solid solubility of tin in lead is 19-5 and of lead in tin 2-6 per cent., both at the eutectic temperature of 183 C. Antimony-Lead.-—The antimony-lead liquidus also consists of two curves intersecting at a eutectic point. Early investigators include Stead 4 and Gontermann,5 who determined the eutectic composition to be 12-8 per cent, antim ony a t 247° C. and 13-0 per cent, a t 245° C.,

respectively. Later observers, of whom the most recent are Broniewski and Sliwowski,6 confirm the eutectic competition as lying between 12-7 and 13-0 per cent, of antimony, solidifying at 247° C .^they also confirm the formation of solid solutions of antimony in lead ranging from 0-5 per cent, antimony at room temperature to 2-45 per cent, at the eutectic temperature, and of 11 per cent, lead in antimony at the eutectic temperature. The formation of these solid solutions has been confirmed more recently by Portevin and Bastion,7 using cast- ability methods. Cooling curves taken by the present author indicate the eutectic composition as being nearer 12-7 than 13-0 per cent, anti mony. The antimony-lead diagram due to Broniewski and Sliwowski is shown in Fig. 2. Antimony-Tin —The antimony-tin system has been studied by numerous investigators, all of whom find the liquidus to consist of three branches having very little change of direction, representing the 19 Weaver : Type Metal Alloys separation of three solid solutions. The most recently published papers on this system are by Aoki, Osawa, and Iwasé 8 and by Bowen and Morris Jones.9 Tlie diagram due to the former is given in Fig. 3. In both investigations the antimony-rich solid solution was found to

Fro. 2.—Lead-Antimony System (Broniewski and Sliwowski). contain 0-10 per cent. tin. Aoki, Osawa, and Iwase found the tin- rich solid solution to contain 0-10 per cent, antimony, whereas Bowen and Morris Jones determined the limit of solubility of antimony in tin as 9 per cent. In both investigations the middle section of

Fro. 3.—Antimony-Tin System (Aoki, Osawa, and Iwase). the liquidus curve was found, by means of X-ray analysis, to repre sent the separation of the compound SbSn capable of bolding both antimony and tin in solid solution. Bowen and Morris Jones deter mined the limits of solid solubility as antimony 4 and tin 10 per cent. 13 Weaver: Type Metal Alloys

Aoki, Osawa, and Iwase claim to have found a change in alloys within this range which they attribute to an allotropic transformation. Antimony-Tin-Lead.—Although this investigation involves only the lead-base corner of the ternary diagram, the complete diagram should be considered. This has been investigated by Loebe,10 Cambell and Elder,11 Cambell,12 Heyn and Bauer,13 and more recently by Iwase and Aoki.14 In the first four investigations the liquidus and solidus surfaces were built up by means of thermal analysis, and are in approximate agreement. The crystals separating were assumed to be the same as in the binary alloys. No true ternary eutectic was found, that occurring was said to be composed of lead containing a little tin in solid solution, and one of the antimony-tin solid solutions which were not distinguishable microscopically. The diagram due to Loebe has been accepted generally for more than 20 years, and is repro duced in Fig. 4. The liquidus surface is composed of four areas representing the primary crystal lization of the three antimony-tin solid solutions occurring in the binary antimony-tin system, and of lead. Describing this diagram in terms of the symbols used throughout this paper, BPOV Fig. 4.—Antimony-Tiu-Lead System represents the prim ary separation 0{ the a antimony-ricli solid solu tion (Loebe’s [i constituent), PliSO that of the p solid solution said to contain the compound SbSn (Loebe’s y constituent), RA IF»S th a t of the y tin-rich solid solution (Loebe’s a constituent), and VOSWC th a t of primary lead. This nomenclature was first adopted by Iwase and Aoki, and is preferable to that of Loebe because of the more consecu tive arrangement of the symbols. Considering the internal boundaries of these areas the two grooves VO and STF represent the eutectic separations of lead and the a con stituent and lead and the y constituent, respectively. Along PO there takes place the peritectic reaction a crystals -f liquid —>- [i crystals, and along RS th a t of [i crystals + liquid—> y crystals. Along the groove OS eutectic separation of lead and [i constituent takes place. 0 and S were assumed to be ternary peritectic points of invariant equi librium. The compositions and solidification temperatures as deter mined by the different investigators are give in Table II. 14 Weaver : Type Metal Alloys

T able II. Tin, Antimony, Lend, Temperature, Per Cent. Per Cent. Per Ceut. "0, Loebe . 1911 10 10 80 245 Cambell and Elder . 1911 10 10 80 245 Cambell . . 1012 10 10 80 245 Heyn and Bauer . 1914 10 10 00 242 Loebc . 1911 60-5 30 30-5 191 Cambell and Elder . 1911 57-5 2-5 40-0 189 Cambell . . 1912 57-5 2-5 40-0 189 Heyn and Bauer . 1914 53-5 4-0 42-5 184 The results determining 0 are in agreement, whilst those determining S show considerable variance. All found that segregation

Tf PER CENT F ig. 5.—Lead-Antimony-Tin System. Isothermal Diagram for Liquidus Surface (Iwasé and Aoki). occurs with slow cooling owing to the difference in density between the crystals of the antimony-tin solid solutions and the lead-rich melt. Thus thermal analysis is difficult, and differences in results may be due to whether stirring is or is not employed. Cambell and Elder definitely stated that they did not use stirring, and their results lose some value as concentration of one constituent in the upper part of the melt hinders the completion of true equilibrium. The other investigators do not give experimental details. Iwasc and Aoki quite recently re-examined the whole of the ternary diagram by means of thermal and microscopical investigations, and put forward the following new ideas with regard to the constitution of these ternary alloys. They found that the ternary peritectic point 0 15 Weaver : Type Metal Alloys of earlier investigators (see Eg. 4) is a true ternary eutectic tlie exist ence of which is supported by microscopic examination, using a new series of etching reagents which distinguishes between the antimony— tin solid solutions. This ternary eutectic has the composition anti mony 11-5, tin 3-5, and lead 85-0 per cent., instead of antim ony 10, tin 10, and lead 80 per cent, as found by the earlier investigators, and solidifies at 240° C. instead of 245° C. The composition of the other invariant equilibrium point, the S of Loebe’s diagram, agrees with that of Cambell, viz. antimony 2-5, tin 57-5, and lead 40-0 per cent, lhe existence of a new maximum point »»' on the groove joining the two invariant points was found, having the composition antimony 10, tin 10, and lead 80 per cent, and solidifying at 245° C., but no satisfactory explanation was suggested. A detailed diagram of the isotherms in the lead-base corner, due to Iwase and Aoki, is given in Fig. 5. In their complete ternary diagram Iwase and Aoki introduce two phases, Pj and p2, corresponding with the y SbSn solid solution of previous workers, but this is substantiated only by work on the binary system, quoted earlier, and the authors themselves admit that they have no evidence, either thermal or microscopic, of the actual existence of the two phases in the ternary diagram. The limits of the solid solubilities of the three constituents have also been defined by Iwase and Aoki. Earlier work was carried out on this subject by Morgan and Roberts.15

E xperimental W o r k . The experimental work was divided into : 1. Preparation and analysis of 80 alloys. 2. Thermal analysis by means of direct time-temperature cooling curves. 3. Microexamination of: Slowly cooled ingots, Cast ingots, and Various type cast by industrial type-casting machines and processes. 4. Hardness measurements.

P r e p a r a t io n o f A l l o y s . Series of alloys were prepared with compositions lying on six lines parallel to the antimony-lead side of the antimony-tin-lead plane constitutional diagram having approximately constant contents of 2, 4, 6, 8,10, and 12 per cent, tin, respectively. The purest metals available were employed, Chempur tin (approximately 99-97 per cent, pure), 16 Weaver : Type Metal Alloys

Cookson’s antimony (approximately 09-7 per cent, pure), and special lead (approximately 99-999 per cent, pure) being used. The lead was obtained from Messrs. Locke Lancaster through the kindness of Mr. W. T. Butcher. The alloys were prepared in a graphite crucible, heated in a Davis gas-fired furnace under a flux composed of equimolecular proportions of barium, calcium, potassium, and sodium chlorides. The antimony was melted first, the tin and lead being added, and the melt well stirred. The alloys were cast into a small U-shaped cast-iron mould as described in a previous paper.10 The compositions of all alloys were checked by chemical analysis. A list of alloys, with analyses, is given in the Appendix.

T h e r m a l A n a l y s is . Direct time-temperature cooling curves were taken using a Cam bridge 6-dial vernier potentiometer, the curves being recorded on paper wound on a clockwork drum which made one complete revolu tion per hr. A needle point carried by a vernier scale was moved bv hand along a graduated scale parallel to the axis of rotation of the drum; the needle was set at the reading corresponding to the potentio meter setting, and, when the mirror galvanometer registered zero, a puncture was made in the paper by depressing the needle; the needle and potentiometer were then set to the next reading and another point marked when the galvanometer again recorded zero. A record of the complete cooling curve was built up in this way. 250 to 300 grm. of each alloy were melted in an electric furnace in a special cylindrical fireclay crucible 1 in. in diameter and in. high, the space between the crucible and furnace being well packed with asbestos fibre. The molten alloy was covered with a seal 1 in. deep of high flash-point cylinder oil which prevented oxidation and also formed a heat-insulating covering. A base-metal Alumel Chromel thermocouple was chosen as being the most suitable for the range of temperature to be measured. The junction was brazed to the base of a hole drilled in a solid Nichrome rod, which fitted rigidly into a depression in the centre of the base of the crucible, direct contact between the thermocouple junction and the melt being thus ensured and lag reduced to a minimum. To prevent segregation and also to ensure uniform heat distribution during cooling, the molten alloy was stirred mechanically by a screw shaped paddle made of fused silica which was rotated round the thermo couple by means of a small motor. A speed of 120 r.p.m. was sufficient to prevent the crystals of the antimony-tin solid solutions from rising B 17

P°UTEC HHk , Weaver : T y pe Metal Alloys to the top of the melt. Stirring was used with all alloys, as heat distribution was found to be uneven where this was not employed. Stirring ensured that the thermocouple registered the average tem perature of the whole melt and not that of the metal localized round the thermocouple. The friction drive of the stirring apparatus was adjusted so that slight increase in viscosity at the commencement of the eutectic separation was sufficient to cause slip and stop the move ment of the stirrer, which was then lifted out of the alloy. As the tem perature of the alloy is then approximately stationary removal of the stirrer has no apparent effect on the cooling curve. The arrange ment of the crucible in the furnace, and of the stirring apparatus, is shown in Fig. 6. The thermocouple was calibrated under the exact con ditions used throughout the investigation against the ac cepted molting points of pure antimony, lead, cadmium, bis muth, and tin and also the boiling point of water. Both heating and cooling curves were taken, but the differences ob served between melting and solidifying points did not exceed 0-01 mv. The plotted mean results give an almost linear Fig. O.-ApparatusforThermal lda|o n, 0-1 mv. being equivalent to 2-3° C. Cooling curves were taken from 350° C. The actual time intervals recorded were for drops in temperature equivalent to 0-05 mv. (1*15° C.) and, where possible during arrests, of OOl mv. (0-23° C.). The thermal results are considered accurate to 05° C. Heating curves were also determined on various alloys, but the results again agreed to within 0-01 mv. with those obtained with cooling curves. Heating curves were considered sufficient for the purposes of this investigation. The ingots obtained with the cooling curve experiments were sectioned vertically, polished and etched for microexamination. Figs. 7 and 8 (Plate I) show the distribution of the crystals obtained with and without stirring. 18 P l a t e I.

Fir,. 7.—Cooling Curve Ingot, Alloy 146. F ig. 8.—Cooling Curve Ingot, Alloy 146. Sb 16-1, Sn 8-0, Pb 75-9%. ' With Sb 16-1, Sn 8-0, Pb 75-9%. Without Stirring, x 1-5 Stirring, x 15

19 P l a t e IT.

F ig . 18.—Alloy 139. Sb 6-2, Sn 6-0, Pb 87-8%. F ig. 19.—Alloy 199. Primary 0 (Grey Cube), Primary 8 (Groundwork). Eutectic fl + S. Eutectic jS -f S. Sb 11*8, Sn 12-0, Pb 76-2%.

Fie. 20.—Alloy 177. Pseudo-Binary Eutectic F ig. 21.—Alloy 16]- 8-4-5. Sb 10-2, Sn 10 0, Pb 79-8%. + S. (a White, 0 Grey.) Sb 12-0, Sn 4-Os, H ' Pb 83-95%. Cooling Curve Ingots. Etched Electrolytically in Aqueous 10% HC1. X 150. 20 P l a t e I I I .

F ig. 22.—Alloy 144. Primary a (White), Sur Fig. 23.—Same as Fig. 22. Mould-Cast Ingot, rounded by Peritcctic p (Grey), and Ternary x 850. Eutectic a + p + S. Cooling Curve Ingot. Sb 22-0, Sn 6-0, Pb 72-0%. X 150.

Fig. 24.—Alloy 91. Primary 0 (Cube). Sur- F ig . 25.—Same as Fig. 24. Mould-Cast Ingot, rounded by Binary Eutectic fi + 8 and X 1000. Ternary Eutectic a + ¡3 + S. Cooling Curve Ingot. Sb 14-0, Sn 8-0, Pb 78-0%. X 150. Etched Electrolytically in Aqueous 10% HC1. 21 P l a t e IV .

Fig. 26.—Alloy 169. Primary a (White) Sur Fig. 27.—Same as Fig. 26. Mould-Cast Ingot, rounded by Ternary Eutectic of a -f p + S. x 1000. Cooling Curve Ingot. Sb 161, Sn 4'0, Pb 79-9%. x 120.

Fxc. 28.—Alloy 144. Primarv a (White) Sur- Fig. 29.—Alloy 173. Primary a (White) Sur rounded by Peritectic j3 * (Grey) Ternary rounded by Peritectic /3 (Grey) and Ternary Eutectic a '+ fl + s. Verv Slowly Cooled. Eutectic a + /3 + S. Very Slowly Cooled. Sb 22-0, Sn 6-0, Pb 72-0%. X 80. Sb 15-1, Sn 6-0, Pb 78-9%. X 100. Etched Elcctrolytically in Aqueous 10% HC1. 22 P l a t e V .

Fig. 30.—Alloy 91. Primary jS (Grev) with F ig . 31.—Alloy 173. Primary a (White) almost Binary Eutectic fi + S and Ternary Eutectic Completely Converted to (Grey). Eutectic a j8 4- S. Very Slowly Cooled. Sb 14-0, Sn a + fi + L Cooling Curve Ingot. Sb 15-1, Sn 8 0, Pb 78-0%. x 80. 6-0, Pb 78-9%. X 150.

Fig. 32.—Alloy 127. Primary a Completely Fig. 33.—Alloy 161. Primary a Converted to 8 (Grey). Eutectic a + jS + 8. Eutectic a + S. Cooling Curve Ingot. Sb 15-6, Cooling Curve Ingot. Sb 20-2, Sn 10-1, Pb Sn 2-05, Pb 82-35%. X 120. 69-7%. x 150. Etched Electrolytically in Aqueous 10% HC1. 23 P l a t e V I.

Fig. 34.—Alloy 157. Primary 8 (Groundwork). Fig. 35.—Alloy 158. Mainly a -1- 8 Binary Eutectic a 8 and a + ¿3 + S. Cooling Eutectic. Cooling Curve Ingot. Etched as Curve Ingot. Etched Electrolytically in 10% Fig. 34. Sb 11-7, Sn 2-05, Pb 86-25%. X 120. HC1. Sb 8-0, Sn 2-1, Pb 89-9%. X 200.

Frc. 36.—Stereotype Cylinder Sb 14-4, Sn 5-9, F ig. 37.—Same as Fig. 36. Etched 5% AgNO, Pb 79-7%. Etched as Fig. 34. X 1250. Solution, x 1250. P l a t e VII.

Fig. 38.—Linotype Slug. Sb 12-1, Sn 3-0, Fig. 39.—Same as Fig. 38. Etchcd 5% AgN03 Pb 84-9%. Etclicd clectrolytically in 10% solution. X 1750. HCl. X 1750.

F ig. 40.—Monotype Type. Sb 18-4, Sn 8-75, Fig. 41.—Same as Fig. 40. Etched 5% AgN03 Pb 72-85%. Etched Electrolytically in 10% Solution. X 2200. HCl. X 2200. 25 P l a t e VIII.

F ig . 42.—Ternary Model.

2 6 Weaver : Type Metal Alloys

M ic r o -examination . Tlie alloys were polished with Nos. “ G,” “ M,” and “ 0 ” Hubert s emery paper, and then with Sylvo metal polish on Selvyt cloth. They were etched anodically in a 10 per cent, aqueous hydrochloric acid solution using a 2-v. accumulator and a copper cathode, with a distance of 3 in. between the specimen and the cathode. Each specimen was connected to the accumulator before lowering into the solution and was withdrawn before being disconnected. After rinsing, the specimens were dried in an air-blast.

This method of etching, developed by the author, gives definite distinction between the three phases present in the alloys considered. It is simpler than that- used by Iwase and Aoki, which requires two solutions and which the author did not find to give consistent results. With the electrolytic etch the a antimony-rich phase, whether present as primary crystals or as eutectic bands, remains white. 1 he p SbSn phase, whether present as primary or converted crystals, or as eutectic bands, becomes light or dark grey according to the length of time of 27 TEMPERATURE 'C TEMPERATURE 0 2 2 ---- evr: ye ea Alloys Metal Type : Weaver D-a*t t * a - iD U Q U Fia. 11.—Series II. Tin = 4 Per Cent. Per 4 = Tin II. 11.—Series Fia. F 6 1 1 1 1 1 2 22 20 16 16 14 12 10 8 6 4 ig LIQUID ♦ ♦ LIQUID . 10.—Series I. Tin = 2 Per Cent. Per 2 = Tin I. 10.—Series ù Nt N. T N E C R E P ONV. *NTtM X+ß+t +

2 evr: ye ea Alloys Metal Type : Weaver 6 0 2 4 6 f 20 ifl 16 14 12 «0 6 6 4 F F F ig . . ig F o. . jo 14.—Series V. Tin = 10 Per Cent. 10Per = Tin V. 14.—Series 15.—Series VI. Tin = 12 Per Cent. Per 12 = Tin VI. 15.—Series MONY.PER CE T EN C R E P . Y N O IM T N A NtOY E CENT ANTtMOHY, PER 30 LIQUID +¿3 LIQUID LIQUID ♦ a lD ü Q d *fl Weaver : Type Metal Alloys etch. Blue or green colour effects are obtained if the time of etch is too short. The lead-rich S phase is either dark grey or black. This dark-phase etch is of the nature of a powder deposit, which can be removed by a very light rub on Selvyt cloth. In the accompanying photomicrographs this lead-rich S constituent was made light where £S was present without a (sec Figs. 18, 19 and 20 (Plate II)), but was left dark where both a and p were present together (see Figs. 22 and 24 (Plate III)). This procedure was adopted to allow for loss of tone during reproduction for printing. The three phases are easily dis tinguishable on ordinary examination both in coarse and in very line structures. R esults of T herm al An a ly sis. Typical cooling curves are shown in Fig. 9, and the arrest tem peratures obtained with all alloys are given in the Appendix. Figs. 10, 11, 12, 13, 14, and 15 show sections of the solid ternary constitu tional diagram built up from the thermal results obtained with the series of alloys containing 2, 4, 6, 8, 10, and 12 per cent, of tin. The arrest points caused by the liquidus, peritectic and binary complex separations were all sharply defined, but those due to the separation of the ternary eutectic were not so clearly marked. Cooling curves taken with an even slower rate of cooling failed to improve the definition of the ternary arrest. The contours of the liquidus surface in the sectional diagrams determine accurately the location of the grooves bounding the different areas of primary crystallization. Cooling curves were also determined for alloys having intermediate compositions, in order to define more exactly the temperature gradient of these grooves. The surfaces below the liquidus are built up from thermal results con firmed by microscopic examination of cooling curve ingots. It is recognized that the solidus is not that which would be obtained with true equilibrium conditions. It is considered that the solidus given is of more practical use, as it represents the results obtained with all practical rates of cooling, including those of industrial methods of casting as described later. Experiments of several weeks’ annealing are necessary to fix the limits of these surfaces under conditions of true equilibrium. The plane isothermal diagram for the liquidus surface is given in Fig. 16. Comparison with that given in Fig. 5 (p. 15) shows that the modifications introduced by Iwase and Aoki are substantially con firmed. The greater detail of the present author’s investigation, how ever, has enabled the contours and boundaries of the liquidus surface to be determined more exactly. The liquidus areas represent the same primary crystallizations, and the grooves the same peritectic and 31 Weaver : Type Metal Alloys binary complex separations as in Iwase and Acid's diagram, and also in that of the earlier workers typified by Loebe. AEXFD represents a portion of the area representing the primary separation of the a antimony-rich solid solution, FXMG a portion of that of the |3 solid solution consisting mainly of the compound SbSn, EXMGGB a portion of that of the S lead-rich solid solution. Along EX separation of tlie binary eutectic complex a + S takes place and along XG that of the binary eutectic complex of [i -|- S. The groove EX is markedly cusped. Along FX the peritectic reaction a + liquid —-> ¡3 takes place.

F ig. 16.—Liquidus Surface.

The occurrence of a maximum point at M having the composition antimony 10, tin 10, and lead 80 per cent, (the in' observed by Iwase and Aoki) is confirmed. This alloy solidifies at constant temperature at 216-5° C., giving a microstructure exactly similar to that of a binary eutectic as shown in Fig. 20 (Plate II). It would thus appear that this point is the eutectic alloy of a pseudo-binary system of lead and the compound SbSn, represented by the dotted line BH in Fig. 16. The thermal results obtained with alloys having compositions on a line joining A, the apex of the lead corner of the ternary diagram, to the 32 Weaver : Type Metal Alloys

SbSn peritectic point of the binary antimony-tin system (antimony 50, tin 50 per cent.) are shown in Fig. 17. This diagram is essentially similar to that of a simple binary eutectic system, and the dotted line B ll in Fig. 16 continued to the SbSn binary side may be said to divide the ternary system into two parts. Bach of these parts may be con sidered as a ternary system, so that the whole is composed of two ternary systems of which the components are Sb, Sn, SbSn, and Pb, Sn, SbSn, respectively. Microstructures of alloys with compositions on this pseudo-binary line are shown in Figs. 18 and 19 (Plate II); they are typical of true binary alloys.

F ig. 17.—Pseudo-Binary System Pb-SbSn.

As the grooves E X , F X , and MX all slope down towards X this must be a ternary eutectic invariant point, as put forward by Iwase and Aoki; this is confirmed by microscopic examination as shown in Fig. 21 (Plate II). That this is not a ternary peritectic point is also proved by the fact that there is no evidence of a reaction ring round the a phase in the ternary eutectic indicating that (3 must separate primarily with a at the ternary eutectic temperature. The composition as determined by the author is, however, antimony 12, tin 4, and lead 84 per cent, instead of antimony 11-5, tin 3-5, and lead 86 per cent, as put forward by Iwase and Aoki. It would appear, therefore, that the ternary peritectic point observed by Loebe and contemporary investigators in this part of the diagram is really the eutectic point of the pseudo-binary system of lead and the compound SbSn. The existence of the true ternary eutectic point was c 33 Weaver : Type Metal Alloys

34 Weaver : Type Metal A lloys missed owing to inadequate methods of thermal analysis and o£ etching for micro-examination. A wire space-model of the constitutional diagram of the lead-base corner containing up to 24 per cent, antimony and 14 per cent, tin has been constructed, and a photograph of this is shown in Fig. 42 (Plate VIII). Figs. 43, 44, and 45 show diagrammatic drawings in perspective of the various surfaces. For the construction of the model the limits of the S solid solution have been taken from Iwase and Aoki's results. Microstructure. The microstructurcs obtained with slow cooling during thermal analysis were compared with those of ingots of the same alloys cast in the small cast-iron mould. The structures of the cooling curve ingots were always much coarser than those of the mould ingots, but the electrolytic etch showed that the micro-constituents were essentially the same. This is shown by a comparison (allowing for the different magnifications) of the photomicrographs, Figs. 22 and 23 (Plate III), showing primary a (white) crystals surrounded by secondary ¡3 (grey) in a matrix of ternary eutectic of a + p + &. This alloy (No. 144) contains insufficient tin to convert all the primary a into p. Figs. 24 and 25 (Plate III) and 2G and 27 (Plate IV) also compare the microstructures of cooling curve ingots with those of mould-cast ingots. Figs. 24 and 25 show primary ¡3 crystals (grey) surrounded by binary eutectic of [3 + S and also ternary eutectic of a ¡3 + 8. Figs. 26 and 27 show primary a (white) in a matrix of ternary eutectic. Experiments were also carried out in which alloys were cooled through the solidification range of temperature at an even slower rate than that used in the thermal analyses. The alloy was placed inside a split cylindrical mould, the lid screwed on, and sealed with fireclay. The mould was rotated slowly by a small motor about a diametral axis inside a large case-hardening furnace which was cooling down to room temperature. The mould was placed in the furnace when the temperature was 400° C., and the time taken for the temperature to decrease 200° C. was approximately 10 hrs. The ingot was removed when cold, sectioned, polished, and etched. Perfect mixing was obtained and, although the microstructures were considerably coarser than even in the cooling curve ingots, the constituents were the same as those obtained with the cooling curves and also in the cast ingots. Photomicrographs of structures obtained with this slow rate of cooling are shown in Figs. 28, 29, and.30 (Plates IV and V). These should be compared with Figs. 22, 31, and 24, respectively, making due allowance for differences in magnification. 35 Weaver : Type Metal Alloys

Figs. 31, 32, 33, 34, and 35 (Plates V and VI) show other typical microstructures. Figs. 31 and 32 again show the effect of the peritectic reaction, but with increasing tin content. In Fig. 31 (alloy 173) the primary a (white) is almost completely converted to p (grey), and in Fig. 32, with higher tin content, no primary a remains, the crystals consisting of converted p. With both alloys the matrix consists of ternary eutectic of a [3 4* S. Fig. 33 shows primary a (white) sur rounded by binary eutectic of a + S. Fig. 34 shows primary S (ground work) with binary a + S eutectic and a little ternary eutectic. Fig. 35 represents the microstructure of the eutectic alloy of the 2 per cent, tin series, and shows mainly a + 8 eutectic. Cooling curve ingot structures are shown in these photomicrographs as the coarse structures are preferable for reproduction. The constituents in the cast ingots were essentially the same, except that with alloy No. 173 no primary a remained showing that the peritectic reaction had proceeded even further than in the cooling curve ingot. The increase in surface area of the primary crystals due to decreased size appears to have more effect on the extent of the peritectic reaction than decrease in reaction time caused by more rapid cooling through the solidification tempera ture. A similar effect was noted with other alloys with compositions near the peritectic line. Prim ary p can be readily distinguished from secondary (peritectic) p both in coarse and fine structures by means of its shape. Primary p forms regular cubes, whereas secondary p occurs generally as irregular shaped crystals, considerably smaller than primary p or primary a.

I ndustrially Cast T y p e s. Figs. 36, 38, and 40 (Plates VI and VII) show microstructures obtained in a piece of stereotype cylinder (antimony 14*4, tin 5-9, lead 79-7 per cent.), of linotype slug (antimony 12*1, tin 3-0, lead 84-9 per cent.), and of a monotype typo (antimony 18-4, tin 8-75, lead 72-85 per cent.), respectively. The composition of the stereotype cylinder lies exactly on the groove FX of the constitutional diagram (see Fig. 16), and the microstructure consists of primary p surrounded by a + P + 8 ternary eutectic. The composition of the linotype alloy lies on the primary a side of the a + 8 binary eutectic groove. The microstructure consists of a little primary a, binary a + 8 eutectic, and a J - p + 8 ternary eutectic. The composition of the monotype alloy lies near the peritectic groove on the primary a side, but the microstructure consists of secondary and primary p surrounded by ternary a + P + 8 eutectic. No a crystals were observed, showing that the primary a was completely converted to p. The size of the monotype 3ti Weaver : Type Metal Alloys type photographed was 36 pt. (J in.) x 18 pt. (£ in.). Smaller types were examined but the structures were too fine for satisfactory repro duction. The microstructures, however, were clearly distinguishable and were essentially the same. Figs. 37, 39, and 4-1 (Plates VI and VII) show the same alloys etched with 5 per cent, aqueous silver nitrate solution, which gives a better indication of the shape of the micro-constituents but does not distinguish between the a and fi phases. The microstructures of industrially cast types are thus in agree ment with the author’s constitutional diagram.

H ardness T e st s. Brinell hardness tests were carried out on all alloys on specimens 1 X 1 X | in., cast in a steel mould with a pouring temperature of

TIN. P£ft CENT Fio. 46.—Iso-Hardness Diagram. Brinell Numbers. 50° C. above the melting point of each alloy and a mould temperature of 50° C. A tensile testing machine was used, with a load of 250 kg. and a 10 mm. ball, the duration of the load being 1 minute; this load was found to give a diameter of impression having the recommended 37 Weaver : ' Type Metal Alloys relationship to the diameter of the ball, lhe hardness numbers obtained from these results are given in the Appendix. An iso hardness diagram has been prepared and is given in Fig. 46. Tests were also carried out on a series of ingots of the same com position (antimony 14-5, tin 7-7, lead 77-8 per cent.), cast in the cast- iron mould with varying mould and pouring temperatures. The results are given in Table III, and the variations should be compared with the variations in microstructure which are described in a previous paper.16 These results show that the hardness is little affected by variations in mould and pouring temperatures as used in these experiments.

Table III. Pouring Temperature. Mould ______*_—------^ Temperature, ° C. 200° 0. ‘-’50° 0. 300° C. 350° C. -100' 0. 100 ‘>4-5 24-5 24-0 23-5 23-5 150 ‘>2-0 23-5 24-5 23-5 23-5 200 24-0 24-5 24-5 250 26-0

Scratch hardness tests were also carried out with a 45° pyramidal steel point on cooling curve ingots with large primary crystals, to ascertain the relative hardness of the a and [3 phases. Ihe results given in Table IV, which also includes those of tests on pure tin and lead, show the relative hardness numbers in terms of load in kg. divided by square of width of scratch in mm.

T a b l e IV. Relative Hardness Number. Kg./mm.*. a . 200 U - 2 8 ° Tin .... 13-7 Lead . . • 4-4

These results show that the p phase is considerably harder than the a phase. The iso-hardness diagram shows that, in general, when the antimony content is constant and greater than 14 per cent., addition of tin increases the hardness of the alloy. In the regions adjacent to the troughs the effect appears to be complicated by changes in microstructure. There is a region of maximum hardness adjacent to the trough X (l, just below the pseudo-binary point, with a softer area immediately above the groove on the primary fS side. In the diagram, the dotted lines are obtained by extrapolation of the sectional curves from which the iso-hardness curves were determined. 38 Weaver : Type Metal Alloys

S u m m a r y a n d D is c u s s io n o f R e s u l t s . Alloys ranging from antimony 6, tin 2 per cent, to antimony 22, tin 12 per cent., remainder lead, including those used in the printing industry as “ type metals,” have been investigated by means of thermal analysis and microexamination. The use of a direct-contact thermo couple and a stirring apparatus has enabled the liquidus surface to be determined with a high degree of accuracy, and a plane isothermal diagram for the liquidus surface has been prepared for alloys con taining up to 24 per cent, antimony and 14 per cent. tin. The existence of a true ternary eutectic, as put forward by Iwase and Aoki, is confirmed, but the temperature of solidification is found to be 239° C. and the composition antimony 12, tin 4, lead 84 per cent. The ternary peritectic point of Loebe and contemporary workers at the composition antimony 10, tin 10, lead 80 per cent., is shown to be the eutectic point of a pseudo-binary system of lead and the com pound SbSn, which divides the complete ternary diagram into two sections. The pseudo-binary system lies along a line joining the lead apex of the diagram to the point on the binary system representing the composition of the compound SbSn, i.e. antimony 50, tin 50 per cent. The pseudo-binary eutectic alloy solidifies at 246-5° C-, forming a maximum point on the groove joining the two invariant ternary points. The existence of this maximum point was first observed by Iwase and Aoki, but was not explained. The surfaces below the liquidus have been built up tentatively from thermal results and microexamination of cooling curves only. It is recognized that the true solidus surfaces of complete equilibrium can only be constructed from the results of experiments of several weeks annealing. It is considered that the surfaces given are of more practical use as they describe all practical rates of cooling. A new method of etching has been devised which distinguishes between the a antimony-rich solid solution and the p SbSn phase whether present as primary crystals or as eutectic constituents. Micro- structures of alloys cooled at different rates through the solidification temperature show the same micro-constituents although the size of crystal and coarseness of the eutectic varies with the rate of cooling. Micro-examination of actual linotype slugs, monotype single types, and stereotype cylinders at very high magnification reveals structures in accordance with the constitutional diagram. From these results it would appear that the effect of increase in surface of the crystals due to their smaller size balances that due to the reduction in reaction time. The coarse structures obtained with very slow cooling are 39 Weaver : Type Metal Alloys

essentially tlie same as the very fine structures obtained by rapid chilling in industrially cast ingots. Hardness tests have been made on all alloys. Scratch tests show that the (3 constituent is considerably harder than the a phase. Ball hardness tests have been used to prepare an iso-hardness diagram.

A cknowledgements . The author wishes to acknowledge her indebtedness to the Governors of the Sir John Cass Technical Institute and to Messrs. The Monotype Corporation, Ltd., for providing facilities, and to their respective staffs for their interest and advice; and to the International Tin Research and Development Council for a grant enabling the completion of this work. She also desires to thank Dr. M. L. V. Gayler, Mr. J. Cartland, M.C., M.Sc., and Dr. H. Hey wood for their interest and valuable assistance.

R e f e r e n c e s . 1 Mundey, Bissett, and Cartland, J. Insi. Metals, 1922, 28, 141. 5 Rosenhain an 1 Tucker, Phil. Trans. Roy. Soc., 1908, 209, 89. 3 Stoekdale, J. Inst. Metals, 1932, 50, 2G7. 4 Stead, J. Soc. Chem. Ind., 1897, 18, 200. 5 Gonterraann, Z. anorg. Chem., 1907, 55, 419. 6 Broniewski and Sliwowski, Rev. Met., 1928, 25, 397. 7 Portevin and Bastien, J . Inst. Metals, 1934, 54, 55. 8 Aoki, Osawa, and Iwasć, Kinzoku no Ktnkyu, 1930, 7, 147-160. 8 Bowen and Morris Jones, Phil. Mag., 1931, [vii], 12, 441-462. 10 Loebe, Metallurgie, 1911, 8, 7, 33. 11 Cambell and Elder, School Mines Quarterly, 1911, 32, 244. 12 Cambell, Metallurgie, 1912, 9, 422. 13 Heyn and Bauer, Berlin 1914; Met. Ind. (Lond.), 1919, 25, 1. 11 Iwasć and Aoki, Kinzoku no Kenkyu, 1931, 8, 253-267. 15 Morgan and Roberts, Proc. Inst. Metals Div. Amer. Inst. Min. Met. Eiw„ 1928, 336-348. 10 Weaver. J. Inst. Metals, 1933, 51, 29.

40 APPENDIX.

Thermal Arrests. Analyses. Brinell Series. Alloy. Binary or Hardness No. Liquidus. Perltectic Solidus. Nos. Sb. Sn. Pb. Arrest. I 150 6-0 2-1 91-9 279*5 214 239 16-3 157 8-0 2-1 89-9 265-5 244 240 17*0 172 11-0 2-1 86*9 247-5 240 241 153 11-7 2-05 80-25 247 240 241 17*5 150 11-9 2-0 SO-1 249 240 240 160 13-9 2-0 84*1 267-5 247-5 240 18*0 161 15-0 2*05 82-35 283-5 240 241 18*5 102 18-7 2-0 79-3 306 240-5 243 19 103 20-9 2-0 77-1 318 247-Ö ' 243-5 19 II 101 5-9 4-0 90-1 276 244 240 16-8 105 8-1 4-05 87-85 265-5 244 241 IS-5 10G 10-0 4-0 80-0 252 211 239-5 20-5 178 11-0 4*1 84-3 213 241 238*5 21-5 107 12-0 4-05 83-95 239 22-5 179 12-6 4-1 83-3 244 • 242 239 108 14-0 4-0 82*0 258 243 240 22-5 109 10*1 40 79-9 274-5 242 239-5 23 170 19*1 4-0 70-9 307 214 240 23-5 171 21-0 4-0 74*4 326 243 240 25 III 139 0-2 0-0 87-8 271 245-5 18-5 07 8-1 0*1 85-8 259 245 241*5 20*5 00 10*1 0-1 83-8 249 243 239 23 140 11-05 0*0 82-95 ... 243 238-5 22 1*1 11-7 0-1 82-2 240*5 213*5 211 22 G1 12-3 Ö-!) 81-8 247*5 243 240 22-5 180 13*1 0-1 80*8 251-5 211 238-5 181 13-8 0-1 80-1 255 242 239 2*2-5 112 14*2 6*0 79-8 256 212 240 ... 173 15*1 0*0 78-9 261-5 255 210 23 171 10-0 0-0 78-0 272 253 241 23 1-13 18-1 0*0 75*9 290 253 240 26 175 20-3 0-1 73*0 301 257 210 26-5 111 22-0 0*0 72-0 320 203 239-5 27 IV 119 5-7 8*2 86* 1 268-5 243-5 242 18 120 8-2 7-9 83*9 258 240 245 22 90 9-9 8*3 81-8 246 214*5 242 22*5 170 10-7 8-0 81*3 245 243*5 240 89 11-2 8*0 80-8 240*5 245-5 242-5 21* 5 115 13-6 7-9 78-5 259*5 243-5 241 22-7 91 14-0 8*0 78-0 25S 243 240 27 140 10-1 8-0 75-9 207 242 ■ 239 92 16-8 8-1 75*1 270 241 239-5 2*7-5 93 17-3 7*8 74-9 273*5 ... 239-5 27-5 118 20-1 8*1 71-8 293-5 273-5 238 147 22-1 8-0 69-9 313 274-5 239 28*5 V 128 5-9 10-0 . 84-1 204 212-5 239 15-6 138 7-9 9*5 82-6 255 245 243 20 122 9-2 10*0 80-8 249 245-5 243 24 177 10-2 10-0 79*8 ... 216-5 21-5 137 10-7 9-9 79*4 249 240 244 123 12-0 10*0 78-0 252 215 243 22-5 148 13-85 9-85 7G-3 210-5 245 242 25 125 16-0 10-2 73*8 273 244 240 149 18-2 10-0 71-8 282*5 245 240 28-5 127 20-2 10*1 69-7 289-5 242 239-5 30 150 22*1 10*1 67*9 300 292-5 240 30-5 VI 129 6-2 12-0 81-8 261-5 241 238-5 15*2 151 0-9 11-8 81-3 257 244 240 17-5 130 8-0 11*9 80-1 250 244-5 22 152 9-1 12-0 78*9 245 242 131 10-2 12-0 77-8 247*5 245 242 22 199 11-8 12-0 76-2 252 244 242 22*5 133 14-0 12*0 74-0 205 243 241 25*5 153 15-9 11-9 72*2 277 245 242-5 28*5 154 17-8 11-9 70-3 284-5 215 241 29*5 136 20-5 12-0 67-5 291 239-5 ... 155 21-4 11-75 66-85 297 293 239-5 30*5

I ntermediate p o i n t s . 190 12-0 3-15 84-85 243-5 240-5 191 11-6 5-1 83*3 243 241 192 10-9 7*1 82*2 213-5 239-5 197 10-2 9-6 80*1 246 241 194 9'7 11-1 79-2 ... 245-5 243

V. Bfl2 This paper Is copyright. It may bo reprinted, wholly^or in part, in^the 692 Safc’menta or o^nio.« ex p rW d in this paper, on which written aUcuM.cn may be sent to the Secretary not later than April 1, 1935. Tbis paper wil, net * m“ h°d *

SOME PROPERTIES OF TIN CONTAINING SMALL AMOUNTS OF ALUMINIUM, MAN GANESE OR BISMUTH.*

By P rofessor D. HANSON,t D.Sc., Vic e -Pr esid en t, and E. J. SANDFORD,} B.Sc., Mem ber .

SYNorsis. Aluminium.—Aluminium has a largo effect on tho strength of tin ; 0-5 per cent, increases the strength of pure tin f r o m about 10 to about 5 tons/in.2, while tlio elongation decreases from about 80 to 30 per cent. Further'additions, up to 1-0 per cent., produce no appreciable effect. The improved properties are not permanent when the alloys aro stored in normal conditions, owing to a deterioration of the material which com mences at tho surfaco and spreads slowly inwards : a brittle skm is formed, which cracks when the alloy is bent or otherwise strained. Tho tensile strength is seriously affected in thin sections, and a m a s s of cracks forms on the surfaco of the specimens. Tho core remains ductile for Ion" periods. Rolled alloys deteriorate moro rapidly than similar alloys in the cast condition, but tho latter are not immune. . Manganese..—A method for alloying manganeso with tin is described. Tho effect of manganese on tho strength of tin is only Blight, and is practi cally independent of hcat-trcatmcnt. Manganese is probably soluble in solid tin to only a very small degree at all temperatures. 1 he addition of about 0-10-0-15 per cent, of manganese to tin greatly refines the crystal size at al! temperatures up to the solidus; tho effect of about 0-2 per cent, is much less. With manganese contents exceeding about 0-3 per cent, a fine grain is again produced. The slight variations m tensile strength have been correlated to corresponding variations in crystal size. Bismuth.—Bismuth greatly increases tho tensile strength of tin, from about 1-0 ton/in.1 in tho pure metal to about 4-5 tons/m. with a bismuth content of 4 or 5 per cent. Heat-treatmcnt has little effect on the strength, but alloys heat-treated near the eutectic temperature have low elongations. An explanation of the mechanical properties is suggested, based on the assumption of certain approximate values for the solid solubility; the values deduced are approximately tho samo as those given by Cowan, Hiors and Edwards.*“ Bismuth has a profound refining effect on the srain-sizo of tin, producing much finer grain structures than any other alloying element yet investigated: it is particularly effective m restraining grain-growth at elevated temperatures.

* Manuscript received October 29, 1934. t Professor of Metallurgy, University of Birmingham. t Research Student, Department of Metallurgy, University of Birmingham.

Note to Abstractors and Other Readers.—This paper will bo published, in permanent form, in tho Journal of the Institute of Metals, \o l. L \I , 1935. Reference should accordingly be as follows—J. Inst. Metals, 1935, 56 (Advance copy). 43 Hanson and Sandford : Some Properties of Tin

I ntroduction .

T i i e work described in this paper represents the second part of a systematic investigation of the effects of added elements on tin—the influence of silver, iron, nickel, and copper having been described previously,1 and is part of the research programme of the International Tin Research and Development Council, for whom the work has been carried out and to whom the authors are indebted for permission to publish the results. They are particularly grateful to Mr. I ) . J . Macnaughtan for assistance in various ways, and for his keen interest in the work. Little information is available as to the effects of other elements on tin, and the present investigation takes the form, therefore, of a rather rapid survey over a large field in order to provide metallurgists at the earliest possible moment with the basic facts relating to the principal useful alloys. A complete investigation of each series of alloys in turn would involve considerable delay in dealing with some alloys about which basic knowledge is urgently required. For this reason the further study of some scientifically interesting points has had to be postponed. In the present series the behaviour of aluminium-tin alloys offers some specially interesting problems, while the causes influencing the grnin- size in all the alloys dealt with offer scopc for much further study. It is the intention of the authors and of the International Tin Research and Development Council that these investigations shall be undertaken in due course ; their postponem ent is due solely to the w ant of published information concerning tin and its alloys, and the belief that a broad preliminary survey is at the moment more urgent than the complete investigation of individual alloy systems.

PART I.—T h e I n f l u e n c e o f A l u m in iu m . Aluminium forms a eutectic with tin, the composition of which has been reported by Heycock and Neville,2 Gwyer,3 Lorenz and Plum- bridge,22 Brady,15 Crepaz,23 and Kaneko and Kamiya 24 to lie between 0-5 and 2 per cent, of aluminium, but the most complete investigation appears to be that of Losana and Carozzi,4 who found it to be 0-58 per cent, of aluminium and the temperature of formation 229-0° C. No compounds are formed in the system and no solid solubility of aluminium in tin is reported. According to Kaneko and Kamiya24 the hardness of these alloys is a maximum at 5 per cent , of tin, but decreases suddenly on approaching pure tin. For the present investigation, “ Chempur” tin (99-99 per cent.) and Hoope’s metal (less than 0-1 per cent, impurities) were used. To 44 Containing Small Amounts of Aluminium, &c. facilitate alloying, the aluminium was hot-rolled to a thin strip, cut up and dissolved in molten tin at 500° C. to give a temper alloy containing 5 per cent, of aluminium. Nine alloys containing up to 1 per cent, of aluminium were made from this, and chill-cast from 260° C. into a mould, 10 X 2 X 0-5 in. During cooling the ingots crackled (after removing from the mould) with a clear, bell-like tone, but no visible cracks were formed, and the ingots were subsequently rolled without difficulty. The same phenomenon has been observed in antimony-tin alloys. They were cold-rolled by passes of 0-02-0-1 in. thickness and self-annealed at room-temperature for 15 days before testing. The ultimate tensile strength was determined on specimens having a gauge- length of 2 in. and a cross-section of 0-5 by 0-1 in. Testing was carried out at a constant rate of strain of 04 in./in./minute; the results are given in Table I.

T a b l e I .— Tensile Tests on Aluminium-Tin Alloys : Cold-Rolled Strip 0-1 in. Thick, Self-Annealed at Room Temperature for 15 Days. Aluminium, Ultimate Tensile Strength, Elongation or Per Cent. Tom/in.*. Per Cent. 0 1-04 83 0-1 1-54 46 0-2 2-35 30 0-3 2-51 12 0-4 3-04 If» 0-5 2-84 29 0-R 2-04 22 0-8 2-58 23 1-0 2-73 23 Xole.—All tho specimens, except those of puro tin, developed cracks in the parallel portion daring testing. The addition of aluminium increases the tensile strength of tin from 1-04 to 3-04 tons/in.2 when 0-4 per cent, is present ; at this composition a maximum occurs, which is accompanied by a minimum elongation, and further additions reduce the tensile strength. During the tensile test, all the specimens with the exception of pure tin developed cracks along the extended portion, at right angles to the direction of pulling (and, therefore, at right angles to the direction of rolling). They are illustrated in Fig. 1 (Plate I), and are rather similar in general appearance to the cracks found in over-rolled brass; they were not found in the rolled material until it was subjected to further strain as by pulling or bending. To test the possibility that the cracks were produced during rolling and were opened out subsequently, a small specimen was cut from a strip of the alloy containing 0-4 per cent, of aluminium, at right angles to the direction of rolling, and broken in tension. It showed cracks 45 Hanson and Sandford : Some Properties of Tin again at right angles to the direction of pulling, i.e. in this case, in the direction of rolling. The cracks develop, therefore, only during straining after the rolling is completed. It was found further that all the alloy strips were brittle and that bending in any direction caused cracks to open; this simple test also revealing that the strips consisted of a brittle layer extending from the surface inwards and a core of more ductile material. When the strips were bent, the initial cracks did not extend deeper than the outside layers, and the core failed only after repeated bendings. In order to study the phenomenon further, and particularly to ascertain if the cracking was a property of the alloys in the cast condition, a series of alloys was chill-cast into a mould 0-5 in. in diameter, machined into test-pieces, and tested within a few days of casting. No more precautions against oxidation were taken than in casting the first batch of alloys. The results of the tensile tests are given in Table II.

T able II.— Tensile Strength of Aluminium—Tin Alloys: Chill-Cast Specimens Machined from Bars 0-5 in. in Diameter', Tested within a Feiv Lays of Casting. Aluminium, Ultimate Tensile Strength, Elongation on 2 in., Per Cent. Tons /in.’. Ter Cent. 0 1-40 69 0-1 2-18 49 0-2 3-62 46 0-3 4-18 41 0-4 4 0 3 26 0-5 5-26 27 0-6 4-93 36 0-8 4-88 30 1-0 4-73 29 5 -0 * 4-49 34 * This alloy was cast from a portion of the temper alloy. The tensile strength of pure cast tin is higher than that of cold- rolled and self-annealed material, but this is usual for metals of low melting point. The maximum tensile strength and minimum elongation is now found at 0-5 per cent, of aluminium in the cast alloys, while the tensile strengths are all higher than those of Table I. No cracks were found in the specimens, which doubtless contributed to the enhanced tensile strengths and also shows that the cracking found in the rolled alloys was not an inherent property of the cast ingots. In order to determine the effect of casting conditions, several ingots of the alloy containing 0-4 per cent, of aluminium were cast in different manners, as shown in Table III, some cold-rolled and one hot-rolled. They were tested at several periods and the results are given in Table III. Cracks did not appear on testing the strips immediately after rolling. 46 Containing Small Amounts of Aluminium, &c.

T a b l e III .— Results of Tests on Alloys Containing 0 4 Per Cent, of Aluminium after Various Methods of Casting.

Ultimate Tensile Snrfacc Conditions. Cast. Treatment. Strength, Tons/in.*.

A. None 4-48 No cracks. 17 days at R.T. 2-73 Slight surface cracks. 40 days a t R.T. 2-37 Surface cracks. 40 weeks a t R.T. Cracked badly when bent. B. Nono 4-06 No cracks. 21 days at R.T. 2-38 Badly cracked. C. None 4-00 No cracks. 21 days at R.T. 2-01 Badly cracked. D. Nono 4-12 No cracks. 70 hrs. a t 218° C. 3-50 No cracks. 18 days at R.T. 242 Badly cracked. E. None 4-12 Trace of fine cracks. 70 hrs. a t 218° C. 3-83 No cracks. 18 days at R.T. 2-47 Badly cracked. F. 2 days at R.T. 3-83 No cracks. 17 days at R.T. 315 Trace of cracks around fracture. ! 40 days at R.T. 2-86 Trace of cracks around fracture. 40 weeks at R.T. ... Cracked badly when bent.

In the treatment, “ none” indicates that the alloy was tested within a Jew minutes of rolling. A.—Normal casting practice as alloys of Table I. Cold-rolled. B.—Temper alloy added to tin at 800° C. Cast at 300° C. Cold-rolled. G\— „ „ „ .. 300° C. IX— „ „ „ „ 600° C. „ 000° C. E.—Samo melt as I), cooled to 300° C. and cast. Cold-rolled. F.—Normal casting practice, as alloys of Table I. Hot-rolled. R.T.—Room Temperature.

The highest tensile strength for this alloy (04 per cent, aluminium) was 448 tons/in.2, approaching the value for the chill-cast alloy (4-63 tons/in.2); this seems to confirm that cracking is not an inherent property of the material as cast, but develops later in the history of the alloy. It is associated with a progressive deterioration of the alloy.. The tensile strength of the hot-rolled strip (F) was not decreased so rapidly by standing in air, but eventually, after 40 weeks, it was no better than the cold-rolled material. There is some evidence that cast alloy is less liable to the deteriora tion than rolled alloy; a cast ingot 40 weeks old was rolled cold from 0-5 to 04 in. thick, no cracks being observed at any stage, and the resulting strip, tested by bending immediately after rolling, was quite free from cracks; 15 days later, however, cracks were freely formed 47 Hanson and Sandford : Some Properties of Tin

on bending the strip. The deterioration of the original ingot, if any occurred during a period of 40 ■weeks’ storage, must have been slight, but it cannot yet be assumed that none had occurred. Still further light on the phenomenon was revealed by bending tests made on the broken halves of the chill-cast test-pieces referred to in Table II. Fifty-one weeks after testing, all except pure tin cracked on bending through quite a small angle (10°-20°), but the cracks extended only a short distance into the metal; repeated bending caused the outer layer of the metal to crumble, and small pieces became de tached, but the specimen as a whole was quite ductile and withstood several reversed bends without fracture. The depth of the defective layer was greatest in the alloys containing most aluminium, but was not deep in any alloy. On paring off the surface lightly with a penknife, the defective layer was removed and no cracks were formed on bending. All the specimens were affected less than rolled strip of the same com position tested after a similar period of storage. Several other investigators have previously reported similar phenomena in these alloys. Wetzel 5 found that tin alloyed with 1 per cent, of aluminium became brittle in a few days and could be crumbled to powder. Riche 6 stated that alloys with 15-50 per cent, of aluminium are attacked by water with the evolution of hydrogen, but Moissan 7 could not confirm this, although it was again reported by Anderson and Lean 8 and by Pechaux,9 while the present authors have confirmed it experimentally in the case of the 50 per cent, alloy. Pechaux found that fresh filings of tin-aluminium alloys decompose water, but that old filings are not so. active. It is quite likely that reaction with atmospheric moisture is respon sible for the formation of a brittle layer, but much further work would be required to ascertain the nature of the phenomena concerned. The authors have not had an opportunity to conduct such an investigation, but hope to do so in the future. For the present, aluminium alone appears to be excluded as an improving addition to tin, since the very marked increase in strength which it confers is not permanent, owing to the gradual development of a brittle surface layer. Possibly some antidote for this action may be found, when the improvement conferred by aluminium may prove particularly useful. In view of the serious limitations to the uses of the aluminium-tin alloys, grain-size measurements were not made.

PA R T II.—T h e I n flu en c e of M an ga nese. Little is known concerning the manganese-tin equilibrium diagram. Williams 10 reported that several compounds are formed, and that the liquidus rises steeply from the melting point of tin, forming a monotectic. 48 Containing Small Amounts of Aluminium, &c.

The manganese used for the present investigation contained about 1 per cent, iron, 0-5-1 per cent, silicon, and about 0-5 per cent, aluminium. Several attempts to make a temper alloy failed because of the oxidation of the manganese, but this difficulty was overcome by the use of a flux. The tin was melted in a wind furnace, fused borax added, and when the melt was at 900° C. the required weight of manganese to give a 5 per cent, manganese content was quickly pushed below the flux. In a few minutes all had dissolved, whereupon the flux was thickened by the addition of alumina, and the melt poured into an open chill mould. Oxidation was slight. lo r the preparation of alloys of low manganese content, several fluxes were tried. A lead-borate glass, fluid at about 250° C., was too heavy and intermingled with the metal, while palm oil ignited at 300° C. Finally, resin was found to be suitable, and the following casting practice was evolved. Eesin, in lumps, was added to the molten tin when only a few degrees above the casting temperature, which was taken to be 40° C. above the liquidus, and the manganese-tin temper alloy quickly added. The whole melt, including the resin, was poured into a hot mould, whereupon the flux floated to the top. The result was a good ingot with a clean surface. The ingots, 0-5 in. thick, were cold-rolled to 0-1 in., and self-annealed for 28 days at room temperature. Tensile tests, at a constant rate of loading of 0-4 in./in./minute, were carried out on specimens with a gauge-length of 2 in. after the following heat- treatments: (a) self-annealing for 28 days; (6) self-annealing for 28 days and then annealing for 3 hrs. at 100° C.; (c) self-annealing for 2S days and then annealing for 3 hrs. at 221° C.; (d) self-annealing for 28 days and then quenching from 221° C., after 3 hrs. at that temperature. The results are given in Table IV. The addition of manganese to tin causes only a small increase in tensile strength, which increases from about 1-0 ton/in.2 in pure tin to 1-5 tons/in.2 when about 0-8 per cent, of manganese is added. Annealing for 3 hrs. a t 100° C. has little effect, b u t a t 221° C. causes a slight decrease in the tensile strength of all alloys. Quenching from 221° C. (just below the solidus) has little effect, suggesting that the solid solubility of manganese in tin is low. The results show that, at a certain manganese content, there is a decrease in tensile strength, followed by an increase as further additions are made. In the self-annealed condition, the minimum is at 0-20 per cent., but after quenching from 221° C., it is at 0-30 per cent. This will be dis cussed in the light of the grain-size of these alloys. The influence of manganese on the grain-size of tin has been deter mined after the following heat-treatments: (a) self-annealing for 28 days; (6) annealing for 24 hrs. at 110° C.; (c) annealing for 24 hrs. at p 49 Hanson and Sandford : Some Properties of Tin

T a ble IV.— Tensile Strength of Manganese-Tin Alloys: Cold-Rolled Strip 0-1 in. Thick, Self-A n nealed at Room Temperature for 28 Days.

Manga Ultimate Elonga Ultimate Elonga Ultimate Elonga Ultimate Elonga nese, Tensile tion, Tensile tion, Tensile tion, Tensile tion, Per Strength, Per Strength, Per Strength, Per Strength, Per Cent. Tons/in.*. Cent. Tons/in.*. Cent. Tons/in.’. Cent. Tons/in.1. Cent.

A.A. B. 1!. C. C. D.D. 0 1-05 63* 0-85 48 0-80 50 0-85 62 002 M 0 97 0-94 45 0-96 46 0-90 48 0-05 1-22 87 1-17 86 1-04 34 0-94 72 0-10 1-33 84 1-27 67 114 39 1-39 47 0-15 1-32 76 1-26 87 1-07 19 * 1-28 38 0-20 1-26 56 1-23 72 M l 63 1-25 52 O-.'iO 1-32 48 1-31 51 1-21 44 1-15 43 0-50 1-44 30 1-44 67 1-39 52 1-45 31 0-80 1-51 37 1-51 43 1-46 44 1-55 63 1-00 1-47 55 1-53 41 1-42 37 1-50 43

* Broke outside gauge marks. A.—Self-annealed for 28 days. B.— „ ,, ; annealed for 3 hre. at 100° C. C.— „ „ ; „ „ „ 221° C. IX— ,, „ „ ; quenched from 221° C. 160° C. ; (d) annealing for 24 lirs. at 215° C. The heat-treatments (b), (c), and (d) were carried out after the alloys had self-annealed for 28 days. The specimens, 2 X 1*5 in. in size, were etched in dilute nitric acid and the grains counted by the intercept method, using a travelling microscope, with the results given in Table V.

T a b l e V.— Grain-Size of Manganese-Tin Alloys : Cold-Rolled Strip, 0-1 in. Thick. Manganese, G rains per sq. cm. Per Cent. A. 1!. C. I). 0-02 3,800 2,800 24 5-5 0-05 9,900 9,100 3,700 125* 0-10 25,200 14,400 12,500 100* 0-15 25,600 15,600 12,100 24* 0-20 3,700 2,800 2,000 1,300 0-30 5,600 7,500 4,100 1,900 0-50 18,500 10,600 8,200 4,500 0-80 13,200 10,300 17,200 4,100 1-00 13,600 7,400 4,800 4,100 * These specimens consisted of a few largo grains and an area of very small ones. In such cases, the results are more or less fortuitous. A.—Self-annealed for 28 days. B.— „ „ ,, ; annealed 24 hrs. at 110° C. C.— „ „ ., „ „ „ 160° C. IX— „ „ „ „ „ .. 215° C. The grain-size produced by recrystallization at room temperature is suddenly decreased from 3800 grains/cm.2 when 0-02 per cent, of 50 Containing Small Amounts of Aluminium , &c. manganese is present, to more than 25,000 grains/cm.2 with 0-15 per cent. With alloys containing 0-20 and 0-30 per cent., a larger grain-size is found, beyond which a refinement again takes place. On annealing at 110° or 160° C., a larger grain-size is again found with alloys containing 0-20 or 0-30 per cent, of manganese than with alloys of a greater or lesser percentage. Additions of 0-05 per cent, and more of manganese arrest the large grain-growth found in tin when annealed at temperatures about 160° C. After annealing at 215° C., large grain-sizes are found with manganese concentrations up to 0-15 per cent., although the grain- size of the alloy containing 0-15 per cent, is greater than that containing 0-05 or 0-10 per cent. With 0-20 per cent, and more of manganese, grain-growth takes place at 215° C., but is not of the magnitude found with alloys containing less of the added element. The results given in Table IV show that the tensile strength reaches a maximum in alloys containing 0-10-0-15 per cent, of manganese, decreases when more is present, and increases again as the percentage of manganese further increases. This maximum in the tensile strength occurs in alloys that have a small grain-size after self-annealing, a change from 0-15 to 0-20 per cent, manganese causes a decrease in tensile strength from 1-32 to 1-26 tons/in.2, and an increase in grain-size from 25,600 to 3700 grains/cm.2. After annealing at 100° C., the same altera tion in manganese content produces a change in tensile strength from 1-26 to 1-23 tons/in.2, and in grain-size from 15,600 to 2800 grains/cm .2 (annealing a t 110° C.). A fter annealing a t 221° C. (or 215° C. for the grain-size measurements), the corresponding change occurs with 0'10- 0-15 per cent, of manganese, when the tensile strength decreases from 1-14: to 1-07 tons/in.2 and the grain-size increases from 100 to 24 grains/ cm.2. With the three heat-treatmonts discussed above, a further in crease in manganese content produces an increase in tensile strength, but, after quenching from 221° C., it continues to decrease even though the manganese content increases from 0-10 to 0-30 per cent., and the grain-size (measured on separate specimens) decreases considerably. To elucidate this point, the ends and parallel portions of the actual tensile test-pieces were etched, (Grain-size measurements have not previously been made on the test-pieces because of the unknown effect of straining, even on the ends.) On annealing at 221° C., the strain produced by machining the specimens caused large grains to grow in those containing up to 0-30 per cent, of manganese. This fact, explains, therefore, the continued decrease in tensile strength from 0-10 to 0-30 per cent, of manganese; it illustrates, also, the difficulty of preventing strain in tin and many of its alloys, with subsequent grain-growth on annealing. 51 Hanson and Sandford: Some Properties of Tin

It is apparent, therefore, that anomalies in the curves connecting ultimate tensile strength with composition, are due mainly to variations in grain-size and the prevention of grain-growth found with certain compositions. The variations in strength are relatively small, but the correlation between these variations and the accompanying change in the size of the crystals appears to be quite definite.

PART III.—T h e I n fl u e n c e of B ism u th. Bismuth forms a eutectic with tin, the composition of which has been reported by Kapp,11 Stofiel,12 Lepkowski,13 Andrews and John ston,14 Brady,15 and Endo,16 to lie between 42 and 45-4 per cent, of bismuth, and the temperature of formation between 135° and 140° C. The figures reported for the solid solubility of bismuth in tin show con siderable disagreement. Lepkowski13 found that 5 per cent, of bis muth was soluble at 140° C., but Bucher,17 by electrical conductivity measurements, found a solid solubility up to 14 per cent. On the other hand, le Blanc, Naumann, and Tschesno,18 using similar methods, found a maximum solubility of l -5 atomic per cent., and reported the presence of a compound BiSn7, which is partly dissociated at room temperature. Solomon and Jones,19 using X-ray methods, found that the solubility of bismuth in tin was negligible at room temperature, but Endo 16 reported a figure of 16 per cent, at 139° C. Cowan, Hiers, and Edwards20 published a diagram which shows a eutectoid transformation occurring at 95° C., and a solid solubility of 6 per cent, bismuth at 135° C.; below 95° C. there is no solubility. The composition of the eutectoid is 4 per cent, bismuth. On this account, it was hoped that these alloys would be amenable to heat- treatment, and mechanical tests were carried out with this end in view. Alloys containing up to 9 per cent, bismuth were cast from 280° C. into a chill mould, 10 X 2 x 0-5 in., and cold-rolled to 0-1 in. Those con taining up to 5 per cent, of bismuth rolled perfectly, but w7ith higher bismuth contents they cracked badly before much reduction had been made. As it was possible that free bismuth was present in the alloys w’hich did not roll, an alloy with 6 per cent, bismuth was annealed at 122' C. for 4 days to attain the maximum solubility, quenched and cold-rolled. This ingot also cracked before much reduction had been made. In this connection, Pearson 21 showed that the bismuth-tin cutectic can be deformed in an apparently viscous manner provided that the deformation takes place at a slow rate, whereas the alloy would apparently be classed as brittle. It is possible that alloys of high bismuth content could be deformed slowly, as by extrusion, but rolling, even when light pinches are used, produces cracks. 52 Containing Small Amounts of Aluminium, &c.

The alloy strips containing up to 5 per cent, bismuth were machined into test-pieces having a gauge-length of 2 in. and a cross-section of 0-5 x 0-1 in. Mechanical tests, at a constant rate of strain of 0-4 in./in./minute, were carried out after the following heat-treatments : (a) self-annealing for 13 days at room temperature; (6) self-annealing for 13 days and then quenching from 125° C. after 1 week at that temperature; (c) self-annealing for 13 days, annealing for 1 week at 125° C. and slowly cooling; (d) self-annealing for 13 days, annealing for 3 days at 123° C., then 1 week at 76° C. and quenching. (The purpose of this treatment was to determine the influence of the reported trans formation at 95° C.) The results are given in Table VI.

T a ble VI.— Tensile Strength of Bismuth-Tin Alloys : Cold-Rolled Strip, 0-1 in. Thick, Treated as Indicated.

Bis Ultimate Elonga Ultimate Elonga- Ultimate Elonga Ultimate Elonga muth, Tensile tion, Tensile tiou, Tcnsilo tion, Tensile tion, Per Strength, Per Strength, Per Strength, Per Strength, Per Cent. Tons/in.*. Cent. Tons/in.*. Cent. Tons/in.*. Cent. Tons/in.*. Cent. A. A. B. B. 0. O. D. 1>. 0 1-02 89 0-81 44 0-85 56 0-89 37 1 2-31 66 2-12 64 2-09 54 2-16 76 2 3-23 64 3-04 49 2-98 53 314 45 3 3-94 58 3-90 34 3-85 38 4-10 35 4 4-43 53 4-57 18 4-53 17 4-71 18 5 4-61 47 4-83 7 * 4-56 10 4-69 8

* Broke outside gauge marks. A.—Self-annealed for 13 days. B.— „ „ „ ; then quenched from 125° C. C.— ,, „ ,, ; then annealed at 125° C. D.— „ ,, „ „ „ 123° C., and then at 76° C. Bismuth has a marked effect on the mechanical properties of tin. The tensile strength in the self-annealed condition, increases from 1-02 tons/in.2 for pure tin to 4-61 tons/in.2 when 5 per cent, bismuth is present; at the same time, the elongation decreases from 89 to 47 per cent. Annealing at 125° C. causes a marked reduction in the tensile strength of pure tin; and a slight reduction with up to 2 per cent, of bismuth; it has no effect with higher bismuth contents. The same is true of annealing at 123° C. and then at 76° C.; in fact, there is little difference between the effects of the three heat-treatments (B, C, and D of Table VI). In all cases, the elongation decreases considerably (from 44 per cent, for pure tin to less than 10 per cent, with 5 per cent, bism uth). The fractured test-pieces varied considerably in appearance. In the cold-rolled and self-annealed condition, pure tin shows the charac teristic “ orange-peel ” effect when elongated, but alloys with bismuth 53 Hanson and Sandford : Some Properties of Tin have quite smooth surfaces even after fracture. As is shown later, the srain-size of these alloys is extremely small, so that a smooth surface is to be expected. The amount of “ necking ” decreases as the bismuth content increases, and the alloy containing 5 per cent, bismuth fractures with practically no local reduction of area, although the elongation amounts to 47 per cent. After heat-treatments involving annealing at temperatures about 125° C., the elongation decreases to less than 10 per cent., with 5 per cent, bismuth. The appearance after fracture of specimens quenched from 125° C. is shown in Fig. 2 ( ate ). ie erain-size of pure tin is seen to be very large; with 4-5 per cent, bismuth there is again practically no local elongation, and after this lieat-treatment, very little general elongation. The surfaces of the test-pieces were etched and examined. In the 1 per cent, bismuth alloy, the crystals arc covered with slip-bands, while the fracture is intracrystalline; this specimen is illustrated in Fi«'. 3 (Plate II), in which slip-bands are clearly visible. ^ lhe dark edges of the crystals appear to be cracks but are not so, being merely shadows caused by differences of level of adjacent grains. As the bismuth content increases, the number of visible slip-bands decreases, and intcrcrystalline cracks are found, becoming most numerous with 5 per cent, bismuth. This alloy is shown in Fig. 4 (Plate II). iso slip- bands are seen although some crystals appear to have twinned. In ic photographs it is difficult to differentiate between cracks and shadows, but such a differentiation was easily made by a microscopic investigation. This method of investigation was extended to the self-annealed alloys, but here the grain-size is so small that it is impossible to say how failure occurs. This phase of the work is being pursued further m an investigation of the creep of bismuth-tin alloys. While the above observations on the rolling and mechanical propei- t.ies of bismuth-tin alloys cannot be considered to justify accurate estimates of the solid solubility of .bismuth in tin, certain tentative conclusions may be reached. The behaviour on rolling, together with the shape of the strength-composition curves, suggests a solubility at high temperatures in the region of 5 per cent. The almost constant shape of the strength-composition curves, irrespective of heat-treatment, suggests, that the alloys of the different series were actually in the s a m e constitutional condition at the moment of testing, notwith standing the differences in the treatments to which they were subjected; the results can be accounted for by assuming a solid solubility at room temperature less than 1 per cent., and the rapid precipitation of bismuth at room temperature from the supersaturated solid solution of the quenched alloys. (The marked recrystallization of bismuth-tin alloys at room temperature is consistent with the latter assumption.) lhe 54 P l a t e I.

F ig . 1.—Cracks in Tensile Test-Piecc after Fracture. (1% Al.) X 2.

0% Bi.

1% Bi.

2% Bi.

3% Bi.

4% Bi.

5% Bi.

F ig . 2.—Tensile Test-Pieces of Bismuth-Tin Alloys after Fracture. Quenched from 125° C. X i-

55 P l a t e II .

Fig. 3.—Surface of Fractured Test-Piece. (1% Bi.) Quenched from 125° C. x 20.

F ig . 4.—Surface of Fractured Test-Piece. (5% Bi.) Quenched from 125° C. x 20.

56 Containing Small Amounts of Aluminium, &c. improved mechanical properties in the bismuth-tin series must then be accounted for by the intimate distribution of bismuth particles in the tin matrix. With bismuth contents exceeding the limit of solubility at high temperatures, the presence of larger masses of bismuth in the alloy may be expected, and the brittleness observed in rolling can thus be accounted for. This explanation does not take into consideration the low elongations of heat-treated bismuth-tin alloys, which may, however, be due to variations in the distribution of the constituents which are not inconsistent with the suggested hypothesis. The solid solubilities assumed in this hypothesis are approximately those given in the diagram due to Cowan, Hiers, and Edwards,20 but a careful examination of the solid solubility of bismuth in tin is clearly required, and the authors hope to have an early opportunity to under take the investigation. The grain-sizes of bismuth-tin alloys were investigated on heat- treated specimens, 2 X 1'5 in. in size, cut from strip, cold-rolled from 0-5 to 0-1 in. in thickness. After heat-treatment, the specimens were polished for a short time with a mixture of “ Silvo ” and a 25 per cent, solution of sodium hydrate in water; this removed grease and allowed the etching solution to act evenly. Different etching methods were found necessary for different specimens. Pure tin, after all heat-treatments, was best etched by swabbing with a 5 per cent, solution of nitric acid in water. With alloys containing bismuth, electrolytic etching in hydrochloric acid, with the specimen as the anode, was found success ful; those which had been self-annealed only, were etched in con centrated acid, whereas for those annealed at higher temperatures, dilute acid was used. In both cases a black deposit which formed was wiped off, leaving a good grain-etch underneath. The following lieat- treatments were used : (a) self-annealing for 23 weeks; (b) self-annealing for 23 weeks and then annealing for 2-1 hrs. at 104° C.; (c) self-annealing for 23 weeks and then annealing for 24 hrs. at 130° C. Grains were counted by the intercept method, except with the specimen of pure tin annealed at 130° C., in which case the total number of grains over the whole surface of the specimen was counted. With this exception, a travelling microscope was used for the annealed alloys, and the microscope of a Yickers diamond hardness testing machine for the self-annealed specimens. The results are given in Table V II. After cold-rolling, recrystallization takes place in bismuth-tin alloys at room temperature. The grain-sizc of pure tin is 1400 grains/ cm.2, but the addition of 1 per cent, of bismuth refines the grain to 122,000 grains/cm.2. Further additions cause an additional refining, resulting in 314,000 grains/cm.2 with 5 per cent, bismuth. On annealing 57 Hanson and Sandford : Properties of Tin, 6-c.

T a b l e VII.— Grain-Size of Bismuth-Tin Alloys : Cold-Rolled Strip, 0-1 in. Thick. Bismuth, Grains per sq. cm. Per Cent. A. B. 0. 0 1,400 1,200 100 1 122,000 10,700 1,700 2 231,000 12,500 2,400 3 221,000 12,100 2,200 4 303,000 11,000 2,400 5 314,000 8,800 2,500 A.—Self-annealed for 23 weeks. B.— „ „ „ , then for 24 hrs. at 104° C. C-— „ „ „ „ „ „ 130° C. at 104° C., little grain-growth takes place in pure tin, but considerable growth is found in the alloys containing bismuth. With 1-5 per cent, bismuth, the grain-size averages 11,000 grains/cm.,2 a twenty-fold increase over self-annealing. The grain-size is still very much smaller, however, than that for pure tin. On annealing at 130° C., considerable growth takes place in pure tin, resulting in 100 grains/cm.2. The grains of the specimens containing from 1 to 5 per cent, bismuth arc still much smaller, averaging 2000 grains/cm.2, although grain-growth does take place at that temperature. Bismuth effects a very great refining action on the grain-size of tin and prevents any large grain-growth at high temperatures. The effect of bismuth is greater in this respect that that of any other alloying element which the authors have investigated.

B ibliography . 1 Hanson, Sandford, and Stevens, J. Inst. Metals, 1034, 55, 115. 2 Heycock and Neville, J. Chem. Soc., 1890, 57, 37(5. 3 Gwyer, Z. anorg. Chem., 1906, 49, 315. 1 Losana and Carozzi, Gazz. chim. ital., 1923, 53, 546. I Wetzel, Z. Metallkunde, 1922, 14, 335. 6 Riche, J. pharm. Chem., 1895, 1, 5. 7 Moissan, Ann. Chim. phys., 1896, 9, 337. 8 Anderson and Lean, Proc. Roy. Soc., 1903, [A], 72, 277. s Pechaux, Compt. rend., 1904, 138, 1170. 10 Williams, Z. anorg. Chem., 1907, 55, 1. II Kapp, Ann. Physik, 1901, 6, 754. 12 Stoffel, Z. anorg. Chem., 1907, 53, 147. 13 Lepkowski, Z. anorg. Chem., 1908, 59, 288. 14 Andrews and Johnston, J. Inst. Metals, 1924, 32, 385. 15 Brady, J. Inst. Metals, 1923, 2 8 , 397. 16 Endo, Sei. Hep. TohSlcu Imp. Unin., 1925, [i], 14, 479. 17 Bucher, Z. anorg. Chem., 1916, 9 8 , 97. 18 Le Blanc, Naumann, and Tschesno, Ber. Verhandl. Sachs. Akad. D't'sii. Leipzig. 1927, 79, 71-108 (abstract Chem. Zentr., 1928, 99, I, 401). 19 Solomon and Jones, Phil. Mag., 1931, [viii], 11, 1090. 10 Cowan, Hiers, and Edwards, Amer. Soc. Steel Treat. Handbook, 1929, 563. 21 Pearson, J. Inst. Melals, 1934, 54, 111. 22 Lorenz and Plumbridge, Z. anorg. Chem., 1913, 83, 243. 13 Crepaz, Giorn. chim. ind. appl., 1923, 5, 115. -* Kaneko and Kamiya, Nihon-K6gy6heaishi, 1924, 40, 509 (abstract ./. Inst. Melals, 1926, 36, 436). 68 PAPER No. 693. This paper Is copyright. It may bo reprinted, wholly or In part. In the Press (with due acknowledgment) after being presented at a meeting of the Institute to be held on March 0-7, 1935, in the Hall of the Institution of Mechanical Engincere, Storey• s f ? Q O Gate Westminster, I.ondon, S.W.l. The Institute as a body is not responsible for the U i7«J statements or opinions expressed in this paper, on which written discussion may be sent to the Secretary not later than April 1, 1935. This paper will not be reissued in the form of a separate “ Advance Copy,” a method o! publication which has been discontinued. ALLOYS OF MAGNESIUM. PART II.—THE MECHANICAL PROPERTIES OF SOME WROUGHT MAGNESIUM ALLOYS.*

By W. E. PRYTHERCH.t M.Sc., M e m b e r .

S y n o p s is . An investigation into tho mechanical and rolling properties of some magnesium alloys is described. The alloy systems studied have been selected with reference to their constitution with a view to the production of alloys of good mcehanical properties amenable to heat-treatment. Although no alloys have yet been made which respond satisfactorily to heat-treatment in tho manner characteristic of certain well-known aluminium alloys, somo alloys having interesting properties havo been studied.

I ntroduction .

T h is paper represents the second p arti of an investigation on magnesium alloys, in progress at the National Physical Laboratory, and carried out under the direction of the Alloys Sub-Committee of the Aeronautical Research Committee for the Department of Scientific and Industrial Research. The work is treated under the following headings :— 1. Methods of melting and casting. 2. Methods of rolling. 3. The zinc-magncsium alloys. 4. The cadmium-magnesium alloys. 5. The aluminium-magnésium alloys. 6. The zinc-cadmium-magnesium alloys. 7. The zinc-cadmium-aluminium-magnesium alloys. 8. The cadmium-alununium-magnesium alloys, i). Conclusions. * Manuscript received October 23, 1934. f Scientific Officer, Department of Metallurgy and Metallurgical Chemistry, National Physical Laboratory, Teddington. Î .1. L. Haughton and It. J. M. Payne, J. Inst. Metals, 1934, 54, “ Alloys of Magnesium. Part I.—The Constitution of the Magnesium-Itich Alloys of Mag nesium and Nickel.”

Note to Abstractors and Other Readers.—This paper will bo published, iu permanent form, in the Journal of the Institute of Metals, Vol. I AT, 1935. Reference should accordingly be as follows—J. Inst. Metals, 1935, 56 (Advance copy). 59 Prytherph : Alloys of Magnesium.—Part II.

1. Meth od s of Me l t in g and Ca stin g . Owing to th e ease with which magnesium and its alloys combine with oxygen and nitrogen when exposed to the atmosphere at temperatures above the melting point, considerable difficulty was experienced at first in producing chill-cast ingots suitable for rolling into strip. The details of the methods of melting employed have varied from time to time, but, in general, they have not deviated fundamentally from those employed commercially. The following method was used: The requisite amount of magnesium was melted under flux in an iron crucible in a gas-fired natural-draught furnace. A number of fluxes was tried

iia . 1.—The Effect of Heat-Treatmenfc at Various Tempera tures on the Brinell Hardness of Cold-Rolled Magnesium. but those ultimately preferred were the proprietary fluxes known as “ Elrazel,'’ one being of low and the other of high melting point. After melting under the more fusible flux, the alloying elements were added, the metal thoroughly stirred and allowed to stand for a few minutes. A quantity of high melting point flux was then added, the casting temperature adjusted, and the metal poured. Cast-iron moulds, sprayed with a mixture of French chalk, boric acid and sodium silicate,’ were used, and were preheated to 350° C. for the strip ingots and 250° C.’ for the 3-in. diameter billets for rolling to rod. A stream of sulphur dioxide gas was directed on to the metal and into the mould during casting, and the metal in the crucible was further protected by a closely Pry they ch : Alloys of Magnesium.—Part II. fitting lid carrying a small crucible filled with sulphur immediately underneath it. Ingots of satisfactory appearance and density were obtained in this way. Although their surfaces appeared sufficiently good for rolling, better results were obtained if they were first machined.

2. Method s of R o l lin g . In view of the difficulties encountered in an attempt to roll alloys containing more than 2 per cent, zinc, it was considered desirable to

TIME or ANNEALING. MINUTES Fio. 2.—Tho Effect of Heat-Treatment at Various Tem peratures on tho Brinell Hardness of Cold-Rolled Zinc— Manganese-Magnesium Alloy. (W. 1233.) Composi tion : Zinc, 3-08; Manganese, O'54%.

investigate the rolling properties of magnesium alloys in general. Tests were carried out to determine the variation of hardness with time of annealing atvarious temperatures,the results beingshown in Figs. 1 and 2. The materials used were (a) pure magnesium (99-92 per cent.), and (b) an alloy containing zinc 3-08 and manganese 0-54 per cent. In each case the metal was cold-rolled, giving an 8 per cent, reduction in cross-sectional area. Prolonged annealing of pure magnesium at 150° C. 61 Prytherch : Alloys of Magnesium.—Part II.

does not completely soften the material and most of the effect at any given temperature is produced during the first 10 minutes’ exposure. The curves for the 3-08 per cent, zinc-0-5 per cent, manganese alloy show that, at temperatures of 250° C. and below, precipitation-hardening results from continued annealing and that progressive softening occurs at 300° C. The material is completely softened by annealing for 10 minutes at 350° C. These results are in substantial agreement with those obtained by Archbutt' and Jenkin.*

T a u l e I.

Time of Total Schcdulo No. Number Details of Rolling*! Annealing. of Stages. A1 30 min. 18 1st stage, 2% reduction, increasing by 2% until 8th stago; then 16% per stage. A3 30 min. 15 As A1 to 11th stage, then 1 stago with 25% reduction and 3 stages with 33% reduction. A4 30 min. 14 As A1 to 11th stage, then 33%, and two stages of 40% reduction. B1 2 hrs. 18 As Al. B2 2 hrs. 15 As Al up to 8th stage, then 1 stago with 20% reduction and 6 stages with 25% reduction. Cl 1 hr. 24 10% reduction per stage. C2 1 hr. 17 10% reduction per stago to 12th stage, then 25% reduction per stage. 1)1 1 hr. 10 25% reduction per stage.

t Ono pass was given per stage except in the case of Schedules Nos. Cl and C2 where two passes per stage were given in the earlier stages, and No. i ) 1 which started with four passes per stage, the number being gradually reduced. In order to investigate the effect of varying the rolling procedure on the mechanical properties of the resulting sheet, teste were made on two chill-cast slabs measuring 12 x 5 X 2 in. of a 2 per cent, zinc alloy. Each of these was halved and the surfaces machined. The slabs were then rolled at 300° C. in 8 different ways, the details of rolling being given in Table I. The sheet produced was tested both along and across the direction in which rolling was finished. The results in Table II show that, under the conditions of the test, the tensile results obtained on the sheet do not vary appreciably on varying the manner in which * S. L. Archbutt and J. W. Jenkin, Aeronaut. lies. Cttee. Ji. and il/,, Xo. 1037, 1926. 62 Prytherch : Alloys of Magnesium.—Part II. the metal is rolled at constant temperature. Tlic results of the tests in the transverse are slightly better than those in the longitudinal direction. It must be emphasized that during these rolling tests, care was taken that the sheet did not cool during rolling.

T able II.

Rolling Direction of Test Yield-Stress, Ultimate Stress, Elongation on with Hespect to Tons/in.1. Tons/in.*. 2 in., Per Cent. Procedure. Kolling Direction. 15-0 28-8 Schedulo A1 / longitudinal 106 \transverso 13-3 15-8 22-0 15-0 30-0 Schodulo A3 j longitudinal 11-3 \ trans verse 13-0 16-2 26-2 15-4 31-2 Schedule A4 J longitudinal 10-!) (.transverso 12-5 15-6 29-8 I longitudinal 10-0 15-1 25-0 Schedule B2 \transverse lit» 1.5-4 29-3 15-1 26-3 Schedulo Cl 1 longitudinal 11-7 \ transverse 13-3 15-7 29-8 1 longitudinal 11-0 15-3 28-7 Schedule C2 \ transverso 13-3 160 28-7

In order to determine the effect of rolling temperature on the properties of magnesium alloy sheet, a portion of the 2 per cent, zinc alloy was rolled to a thickness of approximately 0-125 in. according to Schedule D1 (Table I). The sheet was cut into a number of test lengths each of sufficient area to provide two tensile test-pieces. The test lengths were annealed together at 300° C. for 15 minutes, quenched in cold water, and then heated separately to the various temperatures for 10 minutes and rolled at the preheating temperature in one pass to sheet 0-107 in. thick, giving a reduction of approximately 20 per cent. The resulting material was then tested in tension in the direction of rolling, the results obtained being given in Table III and illustrated

T abiæ III.

Temperature of Yield-Stress, Ultimate Stress, Elongation on 2 Brineli Kolling, ° 0. Tons/in.1. Tons/in.*. in., Per Cent. Hardness No. 410 10-4 14-7 22-3 51-6 370 10-2 15-1 22-7 310 11-8 15-1 25-8 52-8 280 12-3 15-5 24-8 55-5 255 13-0 15-8 23-8 57-0 225 14-5 16-5 15-8 64-6 190 14-2 17-1 7-0 68-5 172 16-5 17-0 8-5 60-5 156 no yield 15-7 1-5 127 no yield 2-8 0 65-9

03 Pry they ch : Alloys of Magnesium.—Part II.

111* ‘S- 3- They show that the temperature at which a sheet is finished may exert a profound effect on its mechanical properties. The yield-stress may be increased from 104 tons/in.2 for sheet rolled a t 410° C. to 16-5 tons/in.2 for th a t rolled a t 172° C. The ultim ate stress, Brmell hardness, and percentage elongation fall into three well-defined zones, viz. 400°-300° C.,300°-190° C., and 190° C. to room temperature. Rolling between 300° and 400° C. does not appear to cause any hardening of the material. The elongation is lower in material

Fio. 3.—The Effect of Rolling Temperature on the Properties of a 2% Zine-Magnesium Alloy.

roiled at 410° C. than that rolled at 300° C„ probably owing to the rapid gram-growth at the higher temperature. Rolling at between 300 and 190° C. produces material which is progressively harder as the ro mg temperature is reduced. Within this range, a decrease in rolling temperature is accompanied by an increase in ultimate strength and a corresponding decrease in elongation. The decrease in ductilitv is precisely what would be expected in the cold-rolling of a ductile material such as copper or aluminium, so that it may be inferred that in rolling at temperatures between 200° and 300° C. a 2 per cent, zinc-magnesium alloy behaves m a manner similar to copper or aluminium at room 64 Prytheych : Alloys of Magnesium.—Part II.

temperatures. Rolling below 200° C. causes complete destruction of the material from a mechanical point of view; the ultimate stress and elongation decreasing to very low values. During the rolling of certain magnesium alloys, therefore, extreme care should be exercised. If the finishing temperature be too low, there is a risk of the material being completely ruined. The exact degree of hardness can be controlled by adjusting the finishing temperature; 220° C. was found in practice to be suitable for this particular class of alloy and has been used in the experimental work for producing hard-rolled sheet. An X-ray study of the rolled material used in the above experiments was carried out by Mr. C. Wain wright at the National Physical Labora tory. It was shown that the strip of 2 per cent, zinc alloy studied could ' be allocated to three groups identical with those obtained by study of the mechanical properties. The X-ray observations were as follows : The material was found to be preferentially oriented so that the basal planes of the hexagonal lattice were parallel to the direction of rolling with a scattering of 23°. This observation agrees with that of Schmidt, who found preferential orientation in the same manner with a scattering of 20°. Further, it was found that material rolled between 300° and 400°C. consisted of coarser grains than the other material, which suggests that these temperatures produce rapid grain-growth in this alloy. Strip rolled between 300° and 200° C. gave X-ray patterns which showed that the material was relatively fine-grained and that no distortion of the lattice had occurred in these specimens, b u t strip rolled below 200° C. was fine-grained with the crystal lattice considerably distorted. The strip which X-ray examination showed to possess a distorted lattice corresponded with the material which, on testing, was found to be valueless from a mechanical point of view. Further experiments showed that preferential orientation, once it had been produced in a sheet of ¡jure magnesium (99-92 per cent.) by rolling, could not be eliminated by lieat-treatment (500° C. for 30 min utes), but that this treatment produced considerable grain-growth. It was also shown that lattice distortion produced in a sheet of pure magnesium (99-92 per cent.) by cold-rolling could be removed quickly by short-period annealings at low temperature, 20 minutes at 100° C. being sufficient for this purpose. It would appear from these experiments that the mechanical pro perties of magnesium and its alloys are intimately connected with lattice distortion. It is possible to produce work-hardening in these alloys accompanied by a well-defined improvement in mechanical pro perties, but cold-working to the extent of causing lattice distortion immediately ruins the material. The difficulty of rolling magnesium e ' 65 Pry ¿her ch : Alloys of Magnesium.—Part II. can be connected definitely, therefore, with its small capacity for work- hardening. Since distortion of the lattice can be removed by annealing at temperatures considerably lower than that normally used in rolling magnesium, it would appear that the “ rate of annealing ” factor is of considerable importance, and that for successful hot-rolling the rate of removal of distortion by annealing should exceed the rate at which distortion of the lattice by deformation would occur. The critical temperature (i.e. the temperature at which the rate of removal of distortion is the same as the rate of deformation) appears to be 220° C. in the case of the 2 per cent, zinc alloy, for an amount of deformation equivalent to a 20 per cent, reduction in cross-sectional area by rolling; if the amount of deformation applied exceeds this, the critical temperature would presumably be higher and vice versa. This view concurs with practical data obtained during the rolling of magnesium and magnesium alloys.

3 . T h e Zin c -M a g n e s iu m A l l o y s . The constitution of the magnesium-rich zinc-magnesium alloys has been investigated by Hume-Rothery and Rounscfell,* who showed that zinc forms solid solutions with magnesium up to approximately 6 per cent, of zinc at the eutectic temperature (314° C.), decreasing to about 1-7 per cent, at room temperatures. The mechanical properties of the zinc-magnesium alloys have been studied by Archbutt and Jenkin,f and rather more extensively by Dumas and Rockaert.J The latter authors investigated extruded alloys and found that zinc additions up to 11 per cent, increased the tensile strength of magnesium from about 12 to IS tons/in.2. They also observed a marked improvement in the ductility but were notable to produce improved mechanical properties in these alloys by lieat- treatm ent. In the present work the alloys were made in the form of ingots measuring 12 X 5 X 0-75 in. and 12 x 5 X 2 in. For hot-rolling, these were machined, cropped, and preheated for 18 hrs. in an electric muffle furnace. The results of tests on zinc-magnesium alloy sheet, rolled as outlined above, are given in Table IV, from which it can be seen that zinc does not affect the ultimate strength of magnesium to any marked extent but that the elongation is improved. In rolling, the addition of zinc

* W. Hume-Rothery and E. 0. Rounsefell, J. Inst. Metals, 1929, 41, 119. t S. L. Archbutt and J. W. Jenkin, Aeronaut. Jies. Cttee. 11. and M. No. 1285, 1928. J A. Dumas and F. Roekac-rt, Rev. Aluminium, 1932, 9, 1717-1728. 06 Prytherch : Alloys of Magnesium.—Part II.

»0 « I-• ^ ce ► ° <

Tf Cl o - ‘ÔçîS

ci O —

CO »CÎ 1-4 Cl O 9 0 - < ® 7l 1Ô1Ô o » 0A <

O O o - O c£ CO CC £* t- ^ . O• oi “P-i

r- i-5 Tj* pH DJ to a o

•: d P-< : O ¡r. l<£°„ £ ¿ 2 I ® £.B 3% 0 ; Ss C'N O2 a.S I « ¿ o g-c, o o o *-« 1 1 1 ° 3 s! S g -g S l lO' l l l l So ?g *-ceo a P S « i^p: H O fi 67 PrytherçJi : Alloys of Magnesium.—Part II.

up to 2 per cent, appears to soften magnesium, and a lower working temperature may be used. For instance, pure magnesium requires a minimum rolling temperature of 350° C. but a 2 per cent, zinc alloy may be satisfactorily rolled to sheet at 300° C. More than 2 per cent, zinc makes the alloy increasingly difficult to roll. The mechanical properties in the transverse direction are slightly superior to those in the direction of rolling; this is the reverse of that usually experienced with other metals, and the explanation is not clear. Another remarkable feature is that better mechanical properties, particularly as regards ductility, are obtainable from sheet rolled from a 2-in. thick slab than from that rolled from a 0-75-in. thick slab; this has frequently been confirmed in this investigation. The low value obtained for the elongation of the 1-0 per cent, zinc alloy is due to the rolling temperature being rather low for this particular alloy.

4. T h e C a d m iu m -M a g n é s iu m A l l o y s . The constitution of the cadmium-magnesium alloys has been investigated by Hume-Rothery and Rowell,* who found that cadmium up to 75 per cent, is completely soluble in magnesium at room temperature. In order to determine the mechanical properties of these alloys, strip ingots, 0-75 in. thick, were prepared containing up to 15 per cent, cadmium. The test results for these alloys are given in Table V.

T able V.— The Mechanical Properties of the Cadmium-Magnesium Alloys.

Rolling Ultimate Stress, Elongation on 2 Melt Ko. Cadmium, %. Temperature, 0 C. Tops/In.1. in., Per Cent.

W. 1289 A 1-99 330 13-4 4-5 W. 1289 B 1-99 420 13-8 7-7 W. 1290 A 3-89 330 14-1 7-5 W. 1290 B 3-89 420 13-9 8-8 W. 1291 A 5-57 330 14-3 12-5 W. 1291 B 5-57 420 13-8 9-7 W. 1295 A 10-11 300 13-8 16-5 W. 1295 B 1011 400-120 13-9 18-5 W. 1296 A 15-2+ 300 14-3 19 0 W. 1296'B 15-24 400-420 1-4-1 20-0

The results show’ that cadmium additions do not materially affect the ultimate stress of magnesium. The ductility, however, is consider ably increased, this being particularly marked in alloys containing 10 and 15 per cent, cadmium. Rolling at 400°-420° C. gives rather more ductile and softer material as compared with rolling at 300°-330° C. * W. Hume-Rothery and S. \V. Rowell, J. Inst. Metals, 1927, 38, 137. 6S Prytherch : Alloys of Magnesium.—Part II.

Mechanical tests have been carried out on alloys of this series in the form of rolled rod: (1) hot-rolled from 2-75 to 0-875 in. diam eter; (2) hot-rolled from 2-75 to 0-924 in. diameter, followed by cold-rolling from 0-924 to 0-875 in. diameter which is equivalent to approximately 10 per cent, reduction in cross-sectional area by cold-work. Material (1) is referred to briefly in the tables as “ hot-rolled,” and (2) as “ cold- rolled.” The results of mechanical tests and density determinations are given in Table VI. The ultimate strength of these alloys in the hot-rolled condition is low, but they possess good ductility and are able to withstand a small amount of cold-work; after cold-rolling they are considerably stronger and are still moderately ductile.

T a b l e VI.— The Mechanical Properties of the Cadmium-Magnesium Alloys.

Cadmium Ultimate Elongation Reduction in Melt No. Contents Condition.0 Stress, on 1V/ area, Area, Per Density. by Analysis, Cent. Per Cent. Tong/in.1. Per Cent. W. 1394 5-84 H.-R. 15-5 23-5 25-7 1-83 W. 1394 5-84 C.-R. 19-8 17-5 28-9 1-83 W. 1395 12-27 H.-R. 15-5 24-2 39-0 1-93 W. 1395 12-27 C.-R. 18-8 17-0 24-4 1-93 W. 1383 24-7 H.-R. 10-6 22-7 29-4 2-19 W. 1383 24-7 C.-R. 20-3 150 14-8 2-19

* H.-R. = Hot-rolled; C.-R. = Hot-rolled from 2-75 in. to 0-924 in. diameter, followed by cold-rolling from 0-924 in. to 0-875 in. diameter.

5. The Aluminium-Magnésium Alloys. Of the magnesium alloys, those with aluminium are by far the best known and their mechanical properties have been fairly well established. They have been included in this investigation to serve as a basis of comparison, and also to determine their properties in the cold-rolled condition. Billets, 3 in. in diameter, were prepared and rolled to 0-875 in. diameter as previously described. The mechanical properties of the alloys prepared are given in Table VII. The alundniurn-magnesium alloys withstand hot-rolling up to an aluminium content of 7 per cent., and are sufficiently malleable in the hot-rolled condition to withstand further cold-rolling up to 6 per cent, aluminium. These alloys possess good tensile strengths. The results obtained on the 6 per cent, aluminium alloy (W. 1391) were considered to be exceptionally good and it was decided, therefore, to carry out confirmatory tests. The results on the 7 per cent, alloy were, however, disappointing; this was ascribed to the low rolling 69 Prythercti: Alloys of Magnesium.—Pań II.

T a b l e VII.— The Mechanical Properties of the Aluminium-Mdgnesium Alloys.

Ultimate Elongation on Aluminium, Condition.® Stress, Reduction in Melt. No. Per Cent. ■iVarca, Per Tona/in.*. Cent. Area, Per Cent.

W. 1388 4-14 H.-R. 17-3 19-2 35-0 W. 1388 4-14 C.-R. 22-2 7-0 13-2 W. 1376 5-65 H.-R. 18-2 18-7 27-6 W. 1376 C.-R. 23-1 8-5 18-0 W. 1391 5-97 H.-R. 19-3 15-5 23-0 C.-R. 24-3 10-7 14-3 W. 1387 7-16 H.-R. 19-3 13-0 21-8

* H.-ll. = hot-rolled; C.-R. = Hot-rolled to 0-924 in. diameter followed by cold-rolling from 0-924 to 0-875 in. diameter. temperature, which in subsequent tests was increased to 350° C. for the 6 per cent, alloy and 380°-390° C. for the 7 per cent, alloy. The results obtained in the repeat tests are given in Table VIII, and confirm that the 6 per cent, aluminium-magnésium alloy has particularly good mechanical properties, especially when slightly cold-worked. The higher rolling temperature (380° C.) improves the mechanical properties of the rolled 7 per cent, aluminium alloy.

T a b l e VIII.— Repeal Tests on 6 and 7 Per Cent. Aluminium-Magnésium Alloys (IF. 1408 and W. 1426).

Aluminium, Elongation on Bod Dlam. Condition.® Per Cent. Ultimate Stress, Inch. Nominal. Tons/in.*. ■iVarea, Per Cent.

0-625 H.-R. 6 19-5 14-8 0-625 C.-R. 6 24-0 6-4 0-875 C.-R. 6 23-9 f 9-0 f 0-875 C.-R. ; annealed at 6 22-6 f 12-5 f 100° C. for 2 hrs.

0-625 Rolled from 3 -in. 7 20-8 16-8 diam. billet.

* H.-R. = Hot-rolled 2-75 to 0-875 in. diam eter; C.-R. = Hot-rolled from 2-75 to 0-924 in. diameter, followed by cold-rolling from 0-924 to 0-875 in. diameter rod. f One result . -

6. T h e Z in c -C a d m iu m -M a g n e s iu m A l l o y s . In the first series of alloys of this group, the effect of increasing zinc content was investigated on an alloy containing 8 per cent, cadmium. The material tested was in the form of strip rolled from slab ingots 0-75 in. thick. The results obtained are given in Table IX. 70 Prytherch : Alloys of Magnesium.—Part 11.

T able IX .— The Mechanical Properties of the Zinc-Cadmium-Magnesium A lloys.

Ultímate Stress, Elongation on 2 in., Nominal Composition. Tons/in.*. Per Cent. Melt No. Longi Cadmium, Zinc, Longi Transverse. Transverse. Per Cent. Per Cent. tudinal. tudinal. W. 1300 8-0 0-5 14-9 15-5 17*7 25-7 W. 1301 8-0 1-0 14-2 15-4 10-5 14-0 W. 1302 8-0 1-5 14-7 15-2 19-5 22-0 W. 1303 8-0 2-0 15-3 25-0 W. 1304 8-0 2-5 14-9 14-8 W. 1305 8-0 3-0 15-2 9-0

The results of mechanical tests (Table IX) show that although the alloys are not particularly strong they possess a high ductility, which improves with increasing zinc content up to 2 per cent, zinc; with more than this amount they are difficult to roll. It was considered that the alloys were sufficiently interesting to justify further work. A number of alloys was extruded to strip 3 in. wide and 0-375 in. thick, which was then hot-rolled to a thickness of 0-1 in. and tested. Test results on the strip in the extruded and in the extruded and rolled conditions are given in Table X.

T a ble X.— The Mechanical Properties of the Zinc-Cadmium-Magnesium Alloys.

Nominal Composition. Melt Direc Ultimate Klongation on Condition.0 tion of Stress, in., Per Cent. No. Cad> Terit.f Tons/in.*. mium, Zinc, Per Per Cent. Cent. E .7 4 4 E. L. ICO 26-7 E. 8 3 3 E. L. 14-8 25-0 E. 13 2 4 E .. L. 15-6 25-0 E. 7 4 4 E. and H.-R. L. 19-9 (broke O.G.M.) T. 21-0 20-0 E. and H.-R., L. 20-1 {broke O.G.M.) then C.-R. T. 21-8 17-7 E. 8 3 3 E. and H.-R. L. 18-9 9-0 T. 19-7 19-2 E. and H.-R., I,. 19-3 4-0 then C.-R. T. 200 18-5 E, 13 o 4 E. and H.-R. 1>. 20-0 9-0 T. 21-1 19-5 E. and H.-R., L. 20-2 7-5 then C.-R. T. 21-9 11-5

* E. = extruded; H.-R. = hot-roiled; C.-R. = cold-rolled, j* L, s=s longitudinal; T. = transverse. Prytherch : Alloys of Magnesium.—Part II.

Selected compositions of this series of alloys were also rolled to 075 in. diameter rod from 3 in. diameter billets. The rod was tested in the hot-rolled condition and also hot-rolled followed by about 10 per cent, cold-work. The results of the mechanical tests are given in Table X I.

T a b l e XI.— The Mechanical Properties of the Zinc-Cadmium-Magnesium Alloys.

Norn [nal Com pc*sition. Ultimate Elongation Reduction Melt No. Con Den dition.0 S tress, on 4 Varea. in Area, Cad Tons/in.1. Per Cent. Per Cent. sity. mium, Zinc, Per Per Cent. Cent.

W. 1365 8 2 H.-R. 15-6 23-5 38-8 W. 1365 8 2 C.-R. 19-8 12-2 29-0 j 1-89 W. 1372 3 3 H.-R. 15-8 21-5 27-8 W. 1372 3 3 C.-R. 22-1 13-5 21-0 ! 1-83 W. 1377 4 4 H.-R. 17-2 22-0 27-0 \ W. 1377 4 C.-R. 23-8 1-86 4 10-5 21-0 * H.-R. = Hot-rolled from 2-75 to 0-875 in. diameter; C.-R. <= Hot-rolled from 2-75 to 0-924 in. diameter and then cold-rolled from 0-924 to 0-875 in. diameter.

The zinc-cadmium-magnesium alloys in rod form are ductile in the soft condition and possess satisfactory strength and ductility when hard-rolled. It was found that superior results were obtained on certain alloys of this series after cold-rolling if they were quenched immediately following the last pass of hot-rolling. This treatment presumably prevents artificial ageing of the alloy during slow cooling following hot- rolling as a result of which the alloy becomes brittle and the cold-rolled material is poor mechanically. Generally speaking, the alloys of this series show useful mechanical properties since, in the hot-rolled or ex truded condition, they possess a high ductility whether in the form of strip or rolled rod. On cold-rolling the strength is very much increased without serious loss of ductility, and it is possible that these alloys would find application for stampings or forgings and parts which require deformation during fabrication. The hot-working temperature is considerably lower than for the aluminium-magnésium alloys.

7. T h e Z in c -C a d m iu m -A l u m in iu m -M a g x e s iu m A l l o y s . Many alloys of this series were prepared and examined, but only one alloy may be said to possess mechanical properties of practical interest. This contained cadmium 8, zinc 2, and aluminium 6 per cent. 72 Prytherch : Alloys of Magnesium.—Part II.

It is difficult to hot-roll since the operation must be conducted below 330° C., at which temperature the alloy is rather hard. The results of tests 011 this alloy are given in Table XII.

T able XII.— The Mechanical Properties of a Magnesium Alloy Con taining Cadmium 8, Zinc 2, Alum inium 6 Per Cent.

Ultimate Stress, Elongation on Condition. Tons/in.*. 2 in., Per Cent. Density. Hot-rolled .... 20-0 21-9 1-93 Hot-rolled, followed by cold- rolling .... 25-0 10-5 1-93

The high ultimate strength of this alloy coupled with a good elonga tion would suggest th a t this alloy might find a certain am ount of practical application.

8. T h e Cadm ium -A lu m ix ium -M agxesium A lloys. The mechanical properties of a number of alloys of this series are given in Table XIII. Two alloys of this group, viz. W. 1398 and W.

T able XIII.— The Mechanical Properties of the Cadmium—Aluminium- Magnesium Alloys.

Nominal Composition. Ultimate Elongation Réduction ilelt No. Condition.6 Stress, on 4Varea, in Area, Cad Alu- Tons/in.s. Per Cent. Per Cent. mium, miuium, Per Per Cent. Cent.

W. 13S8 4 / H.-R. 17-3 19-2 35-0 4 \ C.-R. 22-2 7-0 13-2 W. 1382 "•i 4 I H.-R. 17-8 20-2 36-5 4 4 \ C.-R. 22-7 9-7 16-5 W. 1384 25 4 j H.-R. 21-1 17-7 2C-0 25 4 1 C.-R. 24'2 5-2 4-2 W. 1376 6 ("H.-R. 18-2 18-7 27-6 6 \C.-R. 23-1 8-5 18-0 W. 1398 " s 6 i H.-R. (380° C.) 19-4 20-0 8 6 \ C.-R. 25-1 9-6 25 6 fH.-R,(3S0°C.) 22-5 20-4 W. 1400 25 6 { H.-R. (300°C.) 23-9 12-8 25 6 1 C.-R. 26-8 7-2 35 6 ( H.-R. (380° C.) 22-6 240 W. 1401 35 6 ■ H.-R, (300° C.) 23-4 200 35 6 1 C.-R. 25-6 15-6 W. 1399 8 8 H.-R. (300° C.) 21-8 13-2

* H.-R. (300° C.) = Hot-rolled a t 300° C.; C.-R. = Hot-rolled followed by cold-rolling. 73 Prytherch : Alloys of Magnesium.—Part II.

1399, containing cadmium 8, aluminium 6, and cadmium 8, aluminium 8 percent., respectively, were considered to possess mechanical properties of sufficient interest to warrant further investigation. It has been pointed out earlier in this paper that the cold-rolled alloys of magnesium may be improved in general value by prolonged annealing at low tem peratures. This treatment was applied in the first instance to increase the limit of proportionality, but no marked increase resulted from the treatment. The heat-treatment, however, benefited the alloys by considerably improving the ductility without detriment to other properties.' Table XIV gives results of tests carried out on these alloys by the Engineering Department of the National Physical Laboratory , the properties of a 6 per cent, aluminium alloy are given for comparison.

T able XIV. .The Mechanical Properties of (1) 6 Per Cent. Aluminium - Magnesium Alloy and (2) 8 Per Cent. Cadmium, 8 Per Cent. Aluminium-Magnesium Alloy.

Nominal Reduc 0-1 % Komina! Composition. Ulti Elonga Limit of tion on tion in Proof Propor Young’s mate Stress, Condition. Stress, tionality , Per Per Tons/ Tons/in.’. lb./in.* Cad Alu Tons/ in.*. X 10-*. mium, minium, in.1. Cent. Cent. Per Per Cent. Cent. 3-3 6-6 6 Hot-rolled 190 17-0 22-8 11-21 21-6 14-0 24-0 13-82 4-39 6-4 8 8 5-19 6-2 0 Hot-rolled, followed 23-S9 8-0 150 19-33 by cold-rolling 26-45 4-5 5-4 20-83 7-38 6-2 8 8 16-52 5-57 6-2 6 Cold-rolled and 22-53 11-5 21-1 annealed 2 lira, at 100° C. 18-36 8-25 6-2 8 8 4 hrs. at 100° C. 25-33 12-0 17-0

These results show that the alloy containing cadmium 8 and aluminium 8 per cent, possesses properties in the cold-rolled, heat-treated condition which compare favourably with those of high-strength aluminium alloys. 9. Con clusion s. Magnesium and its alloys are extremely sensitive to rolling conditions, particularly to rolling temperature. Rolling at too low a temperature causes lattice distortion, which is accompanied by inferior mechanical properties. X-ray investigation shows that magnesium alloy sheet is preferentially oriented so that the basal planes of the hexagonal lattice lie parallel to the direction of rolling. Preferential orientation is not removed by heat-treatment at 500° C. for 30 minutes, but rapid grain-growth occurs as a result of this treatment. Lattice distortion 74 Prytherch : Alloys of Magnesium.—Part II. may be removed from cold-rolled magnesium sheet, however, by short- period annealing at 100° C. In certain magnesium alloys it is possible to produce work-hardening effects by rolling between certain ranges of temperature; the work- hardening so produced is accompanied by improved mechanical pro perties. The results of tests on wrought alloys show that zinc improves the strength of magnesium by a relatively small amount only ; the ductility, however, is markedly improved. Alloys containing up to 2 per cent, zinc are soft and are easily rolled; with greater amounts of zinc hot- rolling becomes increasingly difficult as the zinc content increases. The binary alloys of cadmium and magnesium are soft and ductile, and withstand a small amount of cold-work with improvement of their properties. They are superior in this respect to most of the other alloys investigated. In the cold-worked condition the 6 per cent, cad mium alloy possesses properties comparable with those of the 6 per cent, aluminium alloy in the hot-rolled condition. It is suggested that the cadmium-magnesium alloys would be suitable for hot-forging or pressing. The addition of aluminium to magnesium increases its strength and ductility; the optimum percentage addition of aluminium appears to be between 4 and 7. With 4 per cent, aluminium the alloys are moderately strong and ductile and with 7 per cent, aluminium the strength is good but the ductility is low enough to cause difficulty in working. A satisfactory compromise appears to be about 6 per cent, aluminium. The simultaneous addition of cadmium and zinc to magnesium results in alloys which are moderately strong and possess a high degree of ductility; they may be cold-worked slightly and in this condition possess high strength coupled with a satisfactory ductility. It is suggested that they would be useful in sheet form for hot stamping and pressing. The addition of aluminium to the cadmium-magnesium alloys increases the strength ; addition of cadmium to the aluminium—mag nesium alloys increases ductility and, in certain instances, malleability. The quaternary alloys of cadmium, zinc, and aluminium with magnesium have valuable properties at room temperature but are weak above 300° C. Their working properties (hot or cold) are not good, but it is anticipated that they will extrude satisfactorily. An alloy con taining cadmium 8, zinc 2, and aluminium 6 per cent, has \aluable mechanical properties, particularly when slightly cold-worked. Cold-rolling by a small amount (10 per cent.) improves the mechani cal properties of certain magnesium alloys, and their value is further 75 Prytherch : Alloys of Magnesium.—Part II. increased by low-temperature annealing (100° C. for 4 hrs.); an alloy (cadmium 8, aluminium 8 per cent.) treated in this way (cold-rolled and annealed) possesses mechanical properties comparable with the high- strength aluminium alloys.

A cknowledgments . The author’s thanks are due to the Aeronautical Research Committee and the Department of Scientific and Industrial Research for permission to publish this paper, and to Messrs. British Maxium, Ltd., and Messrs. James Booth and Company (1915), Ltd., for assistance in connection with the work. The author also wishes to express his thanks to Dr. J. L. Haughton, under whose supervision the work was carried out, for his help and guidance throughout, and also to various members of the staff of the Department of Metallurgy and Metallurgical Chemistry for their valuable co-operation.

76 MEETINGS OF OTHER SOCIETIES

THURSDAY , January- 24. R o y a l I n st itu t io n a nd B r it ish Sc ie n c e G u il d .— D r. C. H. Dcsch : " The Microscope I n st itu t io n o f E l e c tr ic a l E n g in e e r s , Scot and the Metal Industries.” (The Royal In t is h Ce n t r e , s t u d e n t s ’ se c t io n .—C. H . stitution, 21 Albemarle St., London, W .l, at Dye : “ The Rectifier and its Application to Electric Welding and Design.” (Royal 9 p.m .) Technical College, Glasgow, at 7.30 p.m.) THURSDAY , FEBRUARY 7. FRIDAY , J a n u a r y 25. Society of Chemical Industry, Chemical N o r t h -E ast Co a st institution o f Sh ip ENGINEERING GROUP.— S. R obson a n d P . S. b u il d e r s a n d E n g in e e r s .— P rofessor F. Lew is : 44 Protective Metal Coatings." Joint Bacon, 11. A. McGregor, and W. S. Burn: meeting with Bristol Section. (The University, “ The Relation of Fatigue to Modern Engine at 7.30 p.m.) Design.” (Mining Institute, Newcastle-upon- T h e Ch em ical So c iet y .— D r. C. IT. Desch (dis Tyne, at C p.m.) cussion opened by) : '* Intermetal lie Com pounds.” (The Society, Burlington House, SATURDAY , JANUARY 26. London, W .l, at 8 p.m.’) INSTITUTE OF BRITISH FOUNDRYMEN, NEW- castle-u po n -T y n e a n d D ist r ic t b r a n c h .— FRIDAY , F e b r u a r y 8. F. Hudson : “ Commercial Moulding Sand Con- ELECTRODEPOSrrORS’ TECHNICAL SOCIETY, BIR t rol for the Tyneside Founder.” (Neville Hall, MINGHAM Se ct io n .— B . J. R. Evans (discussion W estgate R d ., N ew castle-upon-T yne, a t G.15 opened by): '* Metal Cleaning." (James p.m .) W att Memorial Institute, Gt. Charles St., I n st itu t e o f B r it is h F o u n d r y m en , W a les Birmingham, 3, at 7.30 p.m.) and Mo nm o uth B ra n ch .— J . J. McClelland: “ Loam Moulding.” (Merchant Venturers’ Technical College, Bristol, at 6.30 p.m.) S A T U R D A Y ^ February 9. I n st itu t e o f B r it is h F o u n d r y m en , Sc o t o sh WEDNESDAY, JANUARY 30. B ran ch.—A. Marshall: “ Patteminaking.” So c iet y o f Co n su lt in g Ma r in e E n g in e er s (Royal Technical College, George St., Glasgow, a n d Shipbuilders , Sc o ttish D istr ic t.— a t 4 p.m .) Dr. J. H. Paterson: "The Commercial Test ing of Welds.” (Institution of Engineers and TUESDAY , February 12. Shipbuilders in Scotland, 39 Elmbank Cres I n st itu t io n o f E n g in e e r s a nd S hipbuilders cent, Glasgow, at 7 p.m.) in Sco tlan d.—Dr. J. O rr: “ Electric Arc Welding in General Engineering.” (39 Elm S A T URDA F , FEBRUARY 2. bank Crescent, Glasgow, at 7.30 p.m.) Institute of British Foundrymen, Scottish Branch, Falkirk Section.—H. Cowan: SATURDAY, February 16. 44 Modern Sand Control in the Foundry.” (Temperance Café, Luit Riggs, Falkirk, at I n st itu t e o f B r it ish f o u n d r y m e n , E ast 6 p.m .) MIDLANDS BRANCH, LINCOLNSHIRE SECTION.— F. Dunleavy : “ Some Non-Ferrous Foundry WEDNESDAY, FEBRUARY 6. Problems.” ' (Technical College, Monks Road, L incoln, a t 7 p.m .) I n st itu t e o f B r it ish F o u n d r y m en , L anca I n st itu t e o f B r it is h F o u n d r y m en , L anca s h ir e B ranch, P r esto n S ectio n .— T . M akem - son : “ Some Impressions of Foundries in the s h ir e B ran ch .—W. Machin and M. Old ham : “ Castings.” (Engineers’ Club, Albert U.S.A.” (Technical College, Corporation St., Preston, at 7.30 p.m.) Sq., Manchester, at 4 p.m.) I n st itu t e o f B r it is h F o u n d r y m en , L o nd on I n st itu t e o f B r it is h f o u n d r y m e n , W a le s BRANCH.—A . Roger : “ Spraying of Cores and a n d Mo nm o uth B ra n ch .— J . J . S h e e h a n : Castings." (Charing Cross Hotel, London, Synthetic Moulding Sands.” (University W.C.2, at 8 p.m.) College, Newport Rd., Cardiff, at 6.30 p.m.)

CORRIGENDUM. Monthly Journal, 1934, Dec. Page 533, line 6 from bottom. For “ water and passed first over stick caustic soda ” read “ a 30 per cent, aqueous solution of caustic potash and passed first over stick caustic potash.”

77 AUTHOR INDEX TO ABSTRACTS

ALK1NS, W. E.» 31. Gibbons, C. H., 28. Manteil, C. L., 21. Shibacv, S. V., 18. Anderson, C. W., 21. Gil, E. J M 38. Masing, G., 38. Sbutt, W. J., 3. Anderson, 0., 39. Gilbert, W. W., 32. Mathers, F. C., 21, 24. Siebei, E., 31, 39. Armstrong, G., 3. Gillette, C. S., 37. Matsuyama, K., 7. Siebei, G., 8. Arrowsmitli, R., 11. Goldmann, F., 39. Mattocks, E. O., 29. Skinner, C. K., 32. Asato, J., 8. Golikov, K., 14. Meigh, C., 30. Slater, T. G., 8. Gränzer, R., 30. Meunier, F., 19. Smith, D. M., 25. Baiersdorf, G., 23. Greenwood, H., 12. Milboum, M., 24. Someren, E. H. S. van, Bailey, B. W., 12. Gregory, G., 4. Miller, R. F., 5. Barber, C. L., 24. Griebel, C., 35. Mocnnicb, L., 30. Sosman, R. B., 39. Barker, Gh IL, 30. GuCTillot, —, 38. Monroe, VT. W., 40. Souder, W., 26. Barth, T. F. W., 1C. Montequi, R., 24. Southwell, R. V., 28. Bassett, IL N., 12. Haefner, R., 34. Morral, F. R., 16. Sparrow, S. W., 10. Bassett, L. G., 26. Hahn, O., 23. Müller, R. W., 17. Spiere, F. W., 27. Beese, C. W., 40. Haie, F. E., 17. Müller, W., 35. Spitta, —, 36. Biran, A. de, 2. Hall, P. R., 40. Murphv, A. J., 12. Stead, H. V., 33. Blouin, E., 2. Hall, W. T., 24. Mutchler, W., 17. Stein, K., 29. Blum, W., 20, 21. Hammond, R. A. F., 22. Stevens, H., 12. Blume, W., 35. Haason, D., 8, 12. Nakajima, M., 22. Stevens, R. W. B., 27. Booth, H., G. Hariba, H., 9. Naumann, E., 20. Stine, W. E., 33. B< relius, G., 38. Harrington, R. H., 13. Nicolini, W., 21. Straumanis, M., 15. Boston, O. W., 32. Hartshorne, X. II., 38. Nielsen, M. F., 30. Strausser, P. W. C., 20, Breckpot, R., 23. Hase, —, 2. 21. Haughton, J. L., 37. Stuart, A., 38. Brenner, A., 20, 21. Ohl, F., 37. Brick, R. M., 7. Heller, G., 6. Stumpf, L. F., 26. ölander, A., 16. Suitin, M., 14. Brindîey, G. W.*, 27. Hermance, H. W., 23. Ollard, E. A., 21. Brown,'\V. J., 26. Herrmann, E., 35. Summers, R. D., 4. Burgers, W. G., 16. Herrmann, L., 15, 31. O’Neill, IL, 11, 15, 28. Sutton, II., 8,12. Burns, J. L., 13. Hertv, C. H., Jr., 9. Owen, E. A., 9. Swamy, S. R., 15. Butler, J. A. V., 3. Herzfeld, K. F., 31. Hidnert, P., 26. Perlitz, H., 15. Tatu, H., 34. Caley, E. R., 24. Hoar, T. P., 18. Perring, J. W., 21. Taylor, W. J., 8. Cart land, J., 12. Hodgson, C. C., 34. Petersen, L. H., 2. Thews, E. R., 30. Cartwright, W., 31. Ilollman, J. I., 3. Phillips, A., 7. Thomas, J. L., 9. Chadwick, R., 12. Holt, M. L., 26. Pickup, L., 9. Thomassen, L., 16. Chapman, D. L., 4. Hope, H. B., 25. Pierson, G. G., 24. Thompson, J. G., 2. Chase, H., 37. Hothersall, A. W., 22. Pirani, ü., 32. Thompson, J. J., 25. Clark, C. L., 28. növeretad, T., 19. Plaksin, I. N., 18. Thum, A., 39. Clyne, R. W., 5. Howahr, W., 31. Poppy, W. J., 5 . Timbv, E. K., 28. Cochrane, AV., 27. Hurd, L. C., 25. Preiswerk, M., 35. Tombeck, D., 38. Conard, M., 6. Prichard, C. E., 24, Tronstadt, L., 19. Cook, M., 27. Irmann, —, 35. Pye, D. R., 36. Trzebiatowski, W., 6. CudroH, L., 25. Irmann, R., 33. Tunell, G., 16. Ishimaru, S., 24. Quinney, H., 4. Davies, R. M., 26. v. Yargha, G., 16. Dehlinger, TJ., 38. Jackson, D. A., 2. Veeastra, W. A., 4, 5. Reimann, A. L., 36. Vernotte, P., 2, 14. Devereux, W. C., 29. Jacquet, A., 38. Reynolds, F., 25. Dobbins, J. T., 25. Jaeger, F. M., 2, 4, 5. Yivian, A. C., ß. Donaldson, J. W., 10. James, E. G., 26. Ricardo, H. 1t., 36. Voilrath, K., 32. Donnay, J. D. H., 16. Jones, J. W., 14. Richards, E. T., 36. Ritchey, J., 40. Waldram, J. M., 19. Domauf, J., 8. Robertson, J. M., 27. Drucker, A. E., 8. Kenneford, A. S., 11. "Walton, A., 3. Dziewulski, IT., 23. KenworthV, L., 19. Robson, S., 36. Wassermann, G., 16. Kenyon, lt. L., 28. Roeser, W. F., 3. Webb, G. F., 21. Röhrig, H., 1, 20, 21, Weiss, F., 35. Eckert, G., 17, 33. Kersten, M., 9. 31. Ehlers, G., 37. Koehler, W. A., 21, 22. Wendeborn, H. B., 39. Rolfe, R. T., 37. Westgren, A., 16. Ehrenberg, W., 27. Koeppel, C., 31. Rosenbohm, E., 2. Evans, II. L., 12. Konobievskii, S. T., 23. Whipp, B., 27. Kopf, E., 39. Ross, M., 25. White, A. E., 28. Rossié, IL, 38. Widemann, M., 29. Famhain, G. S., 11, 15. Kramers, H. A., 6. Royer, M. B., 9. Field, A. J., 32. Krause, H., 32. Wiechell, H. G., 40. Fink, W. L., 7. Krekeier, K., 20, 32. Wilcox, R. L., 13. Fleischer, F., 21. Kroll, W., 2 . Sachs, G., 15, 31. Wilkinson, G. S., 36. Floe, C. F., 22. Krvnitsky, A. I., 29. Saeger, C. M., Jr., 29. Willard, H. H., 25. Föppl, 0-, 38. Küchler, E., 38. Samans, C. H., 15. Willimott, S. G., 18. Fortner, P., 35. Sändera, K., 35. Wilson, J. E., 16. Fox, J. F., 26. Lahr, G., 32. Sanders, J. P., 25. Witte, F., 9. Franzini, T., 4. Landgraf, O., 38. Sandford, E. J., 12. Wogrinz, A., 2 J. Frary, F. C., 37. Lang, M., 30. Sanford, R. L., 6. Wood, E., 8. Freche, H. R., 7. Languepin, J.-E., 33. Sautner, K., 38. Woodward, R. B., 24. Friedmann, W., 37. Larke, E. C., 27. Schaff, C. W., 21. Worden, E. C., 40. Frölich, W., 23. Lavery, J. H., 28. Schmid, E., 8. Wülirer, J., 34. Fromm, IT., 39. Ijosana, L., 38. Schmitt, H., 20, 34. Wunderlich, F., 39. Fühner, H.. 35. l^udewig, J. W., 10. v. Schwarz, (Frhr.) M., Fuller, M. L., 13. Lux, L., 20, 34. 38 39^ Zarges, W., 33, 34. Schwinning, W., 39. Zarubin, N., 14. Gabriel, P., 25. McAdam, D. J., Jr., 5. Seügman, lt., 8. v. Zeerleder. A., 1. 30, Gallego, M-, 24. Macnaughtan, D. J., 10. Serger, H., 21. 32. 7S METALLURGICAL ABSTRACTS (GENERAL AND NON-FERROUS)

Volume 2 JANUARY 1935 Part 1

I.—PROPERTIES OF METALS

fService Characteristics ol the Light Metals and Their A llo y s . ---- ( Amer. Soc. Test. Mat., 1934, pp. 33).—Prepared by Sub-Committee V II of Com mittee B-7 on Light Metals and Alloys of the A.S.T.M. in conjunction with the American Foundrymen’s Association. The following subjects are dis cussed : (I) metallurgical characteristics—compositions and forms available, applicability to fabricating processes (including methods of casting of cast alloys), types of heat-treatment possible, welding; ( 2) industrial requirements of the aircraft, automotive, general structural, architectural, railway equip ment, and household appliance industries; (3) surface protection—painting, oxide coatings, electroplating. In tabular form are given : (1) trade desig nations, (2) nominal compositions, (3) typical mechanical properties of cast and of wrought alloys of aluminium and of magnesium, (4) compositions of ingot aluminium, (5) physical constants of aluminium and magnesium, ( 6) physical properties of magnesium alloys, (7) foundry characteristics of aluminium and magnesium casting alloys, and ( 8) fabricating characteristics of wrought aluminium alloys. A bibliography is included. [¿Vote by Abstractor : It is impossible to do more than indicate the contents of this valuable up-to-date summary of knowledge of light alloys in use in Am erica; especially as the booklet is itself in the nature of an abstract.]—11. B. D. A Simple Means for Distinguishing the Various Qualities of Aluminium. A. von Zcerleder ( Alum inium , 1034, 17, 88-90).— Two tests are made, one a scratch test with an Aldrey needle (Brinell hardness 70-80) and the other a chemical test in which a drop of 20% caustic soda solution is allowed to remain on the polished surface for 10 minutes. In the scratch test pure aluminium and its a llo ys w ith sm all q u an tities of other m etals, e.g. A lu m ati and A n t icorodal, are marked by the needle, whereas the harder alloys such as Duralumin are un affected. The soda test reveals the presence of copper in the alloys by the pro duction of a black spot. For distinguishing hard-worked from annealed aluminium, scratch tests may be made with hard and with annealed Aldrey needles. Examples of the use of the tests are shown in photographs.— A. R . P. ■fOn the Effect of Increasing the Purity on the Properties and Working of Aluminium. H. Rohrig (Aluminium, 1934, 17, 79-84; and (translation) Light Metals Research, 1934, 3, 233-242).—Recent work on the properties of 5 grades of aluminium varying in purity from 99-39 to 99-995% is reviewed and the results aro tabulated. W ith increasing purity the surface of castings becomes smoother and brighter and the crystal structure gradually becomes less complex until finally a purely large-grained polygonal structure is obtained when the impurities are less than 0 -01% ; the purest metal has a Brinell hardness of 13-9 with an electrical conductivity of 38-1 m./ohm. mm.2. The resistance to corrosion by a mixture of nitric and sulphuric acids decreases rapidly with increasing impurities when the metal is in the hard-rolled state, but there is relatively little difference after annealing at 500° C. and quenching. A 2% copper alloy made with pure aluminium after quenching from 590° C. shows a pure polygonal grain structure, but a sim ilar alloy made wdth alumin ium containing 0 -2% iron after the same heat-treatment shows numerous

* Denotes a paper describing the results of original research. | Denotes a first-class critical review'. B 2 Metallurgical Abstracts Vol. 2 irregularly distributed islets, a ternary iron-copper-aluminium constituent, and the grain boundaries are very irregular.— A. R. P. *Study of the Cooling of Metals [Aluminium] by Air. P. Vemotte and E . B lo u in (Airotcchnique (Suppt. to Aéronautique), 1934,12, 25-26).— A brief, illustrated account of methods used to study the heat excliango between a block of aluminium and currents of cold air passed at various air speeds through a passago cut down the axis of the block. Curves showing the variation of the coeif. of thermal exchange with air speed are given for holes of circular and trefoil sections.—J. C. C. ■¡■The Thermal Properties of Aluminium and Their Applications. A. do B ir a n (Light Metals Rev., 1934, 1, (3 ), 38-49).-— T ra n sla te d from Rev. A lu minium, 1933, 10, 2263-2278; 1934, 11, 2311-2332. See Met. AOs., 1934, 1, 161, 225.— R . B . D. The Light-Reflecting Properties of Aluminium and Its Alloys as Affected by Surface Treatment. ---- H ase (Aluminium, 1934, 17, 20-25).— T h e regular and diffuse reflections of light at an angle of incidence of 45° C. from surfaces of aluminium, Pantal, and Polital which have been treated in various w ays, e.g. highly polished, lacquered, Eloxal-treated, pickled, and scratch- brushed, are shown graphically and briefly discussed.— A. R . P. *The Hall and Allied Effects in Cast Bismuth Plates as Affected by the Rate of Cooling. L. Howard Petersen (Proc. Indiana Acad. Sci., 1933, 43, 185- 190).—Tho thcrmomagnetic and galvano-magnetic effects in bismuth plates are influenced by the rate of solidification (structure) of the alloy. When tho mould of liquid bismuth was cooled in a freezing mixture at — 10° C., the expansion on solidification was rapid enough to eject portions of metal.— R . G. »The Magnetic Moment of the Nucleus of Cæsium. D. A. Jackson (Proc. Roy. Soc., 1934, [A], 147, 500-513).—The nuclear magnetic moment of cæsium is found, from spectroscopic evidence, to be 2-75/1838 magneton to within 5%.—J. S. G. T. *The Exact Measurement of the Specific Heats of Metals at High Tempera tures. XVH.— Calorimetrical Retardation Phenomena in the Case of Cerium and Chromium. F. M. Jaeger and E. Rosenbohm (Proc. K . Akad. Wet. Amsterdam, 1934, 37, 4S9—497).— [In English.] Values of the mean specific heat of cerium between 290° and 400° C. and between 450° and 550° C. are found to depend on the preliminary thermal treatment of the metal and its subsequent cooling. It is possible that a real transition temperature exists between 360° and 370° C. This conclusion is confirmed by X-ray analysis at ordinary temperatures. The phenomenon is associated with tho occur rence of so-called “ one-phaso transitions ” as already shown to occur in beryllium and zirconium. A similar type of transition is also shown by chromium and revealed by measurement of the mean specific heat of the metal within various temperature ranges between 400° and 1066° C.— J. T. *The Refining of Metals by Sublimation in a High Vacuum : Chromium, Aluminium, Silicon, Beryllium. W . Kroll (Metallwirtschaft, 1934,13, 725-731, 789).—These metals can be distilled in a high-frequency vacuum furnace. Good separation from impurities can be obtained readily with aluminium. Beryllium can readily be distilled free from all impurities other than aluminium and magnesium, but even so the product is always brittle.— v. G. ^Effect of Cold-Rolling on the Indentation Hardness of Copper. John G. Thom pson (J. Res. Nat. Bur. Stand., 1934, 13, 745-756 ; Research Paper N o . 742).—Specimens of tough-piteh electrolytic copper, commercial oxygen-free copper, and single crystals of copper of different orientations were subjected to severe cold-rolling to determine the effect on the properties, particularly on tho hardness. In all cases the indentation hardness increased to a maximum value which was maintained during subsequent reduction until the hardness determinations became unreliable owing to the thimicss of the specimens. 1935 I.—Properties of Metals 3

No irregularities were encountered cxcept in the case of very thin specimens. The results were confirmed by determinations of tensile strength of some of the specimens and by the application of Meyer’s analysis to some of the data. The effect of severe cold-rolling on the indentation hardness of copper was not m aterially affectcd by the initial hardness of the specimen, the presence or absence of 0-4% oxygen, the change from polycrystalline to single-crystal specimens, or the orientation of the single crystals with respect to the plane of deformation.—S. G. ‘Preparation of Pure Gallium. James I. Hoffman (./. Res. Nat. Bur. Stand., 1934, 13, 665-672 ; Ilcsearch Paper No. 734).— A method is described for the preparation of pure gallium. The principal operations consist in (1) preparing a hydrochloric acid solution of the metal and extracting the gallium, molyb denum , gold, iro n , and th a lliu m , together w ith sm all am ounts of o th er elem ents ; (2 ) precipitating antimony, arsenic, bismuth, cadmium, copper, germanium, gold, mercury, silver, and tin, and most of the lead, molybdenum, and rhenium, with hydrogen sulphide in an acid solution of the ether extract; (3) precipitat ing iron and. thallium with sodium hydroxide and filtering; (4) depositing the gallium electrolytically from the alkaline filtrate; and (5) eliminating the re maining impurities by fractional crystallization of the metal. Indium and lead are the most persistent impurities, but the last traces can be removed by fractional crystallization. Gallium at least 99-999% pure, containing only very faint traces of iron, lead, and calcium, and having a melting ‘point of 29-780 4- 0-005° C., was obtained.— S. G. ‘ Freezing Point of Gallium. Wm. F. Roeser and James I. Hoffman (J. Res. Nat. Bur. Stand., 1934,13, 673-676; Research Paper No. 735).— The tempera ture of equilibrium between solid and liquid gallium was found to be 29-780 ± 0-005° C. Four determinations on 2 different lots of metal (99-999% pure) all yielded the same result. Difficulties ascribed to the undercooling and low thermal conductivity of the gallium prevented a satisfactory determination from ordinary heating and cooling curves. The temperature of equilibrium between the liquid and solid phases of the metal was obtained by measuring the temperature of an intimate mixture of the solid and the liquid. It was found that the presence of the oxide did not affect the freezing temperature, indicating that the oxide is not appreciably soluble in the metal.— S. G. *The Anodic Passivation of Gold. William James Shutt and Arthur Walton (Trans. Faraday Soc., 1934, 30, 914-926).—From oscillograms of potential variations at an anodically polarized gold electrode the maximum limiting current density (c0) for the anodic dissolution of gold and the time of pas s iv a tio n (T) have been measured in chloride, bromide, and sulphate solutions at 25° C. and in AT-hydrochloric acid at 15°-65° C. The coulombs required to passivate the electrode are given by the expression (c — c0)T — K , w here c is the applied current density. In halide solutions c„ and K are approxi m ately proportional to the halogen ion concentration, but in sulphate solutions, if the concentration of the electrolyte is kept constant by vigorous agitation, K corresponds approximately with the amount of electricity required to produce a unimolecular oxide film on the gold surface. A theoretical con sideration of the results leads to the conclusion that anodic passivation takes place finally by the formation of a film of gold peroxide continuous with the lattice structure of the metal.— A. R. P. *The Anodic Passivation of Gold in Chloride Solutions. G. Armstrong and J. A. V. Butler (Trans. Faraday Soc., 1934, 30, 1173—1177).— In u n stirred chloride solutions the times of passivation of gold electrodes are approxi mately proportional to the concentration of chloride ion and are almost unaffected by replacing hydrochloric acid with potassium chloride. For times above 10 seconds (c — ca)T = K (cf. preceding abstract); K is regarded as the amount of electrolysis required to produce a diffusion layer through 4 M etallurgical A b sir acts Vol. 2 which tlio diffusion of chloride ions to the elcctrodc takes place at the constant rate c„. The thickness of the diffusion layer ( 8) in unstirred and stirred solu tion s is 1-5 — 4 x 10 -cm . an d 4 x 10 1 cm., respectively. When other factors rem ain co n stan t K is proportional, and c0 inversely proportional, to 8.— A . R . P . *Tlie Optical Constants oi Polished and Sputtered Molybdenum Surfaces. R. D. Summers (./. Opt. Soc. Amer., 1934, 24, 261-263).— The computed reflectivity of sputtered molybdenum varied with the conditions of prepara tion and was considerably less than for the massive metal. The results on the massive metal were not appreciably affected by polishing under kero sene or while exposed to air.— R. G. *The Magnetic Transformation Point o£ Heavily Cold-Worked Nickel. H. Q uinnoy (J. Inst. Metals, 1934, 55, 229-240 ; discussion, 240-245).— The Curie point of a rather low-grade sample of commercial nickel was found to be 330° C., i.e. much lower than the accepted value for pure nickel. After severe torsional overstrain the Curie point on heating was raised to 365° C. but returned to the original value on cooling. The raising of the point was observed to be less on subsequent heatings, and entirely disappeared after a full anneal. In the discussion C. J . Sm ithdls referred to tho results obtained by Ransley and him self on 99-9% nickel (J. Inst. Metals, 1932, 49, 287) and suggested that the effects observed by Q. were probably caused by impurities which would increase the lattice distortion. Other speakers endorsed this suggestion. In reply, Q. stated that the nickel contained carbon 0-062, silicon 0 027, copper 0-08, iron 0-18, m agnesium 0-08, m anganese 0-30, su lp h u r 0-012, and n ic k e l 9 9 -0 1 6 % ; he agreed that the effects of these impurities would probably account for many of the observed results.—A. R. P. ♦The Exact Measurement of the Specific Heats of Solid Metals at High Temperatures. XV.— A Redetermination of the Specific Heats of Palladium. F. M. Jaeger and W . A. Veenstra (Proc. K . A kail. Wet. Amsterdam, 1934, 37, 280-283).— [In English.] A redetermination of the specific heats of palladium showed that the maxima previously found in the cp — t an d Cp — t curves were due to experimental error. The cp — t curve proves to be nearly a straight line, although a slight increase in the slope above 1125° C. is evident. No indication of an allotropic change is observed. Tho specific heat c is given by: cp — 0-058378 + 0-120548 X 10"!f + 0-258 X 10 3i-, an d the atomic heat by C = 6-2288 -f- 0-12862 x 10'-i -f- 0-27528 X 1 0 "'i2. Tho v a lu e 3R cal. is exceeded for Cp at — 150° C. and for G, at — 120° C.— S. G. »The Action of an External Electrical Field on Hydrogen-Charged Metals [Palladium]. T. Franzini (Atli li. Accad. Lincei (Roma), 1934, 19, 584- 588; C. Abs., 1934, 28, 7098).—A palladium wire, electrolytically charged with hydrogen in a sodium hydroxide solution, was exposed to the field of a proton rectifying transformer at 25 kv. The resistance of the wiro wras measured with a Wheatstone bridge. A positive potential of 15,000 v. under a vacuum (10 1 mm.) caused a very small loss of hydrogen, and a flow of current of less than 0-1 milliamp. A negative potential of the same mag nitude, however, discharged about | of tho adsorbed gas in 15 minutes, and gave a current of 10-12 milliamp., bccause the protons were torn away from the palladium wire, and withdrawn by the high vacuum.—S. G. ♦The Catalysis by Palladium of the Union ol Hydrogen and Oxygen. A New Phenomenon of Contact Catalysis. D. L. Chapman and G. Gregory (Proc. Iloy. Soc., 1934, [A], 147, 68-75).—The mechanism of the catalysis of the union of hydrogen and oxygen by palladium is mainly one of alternate oxidation of metal and reduction of the oxide. Adsorbed hydrogen does not react appreciably with oxygen at room temperature. By suitable treat ment palladium can "be rendered tem porarily inert as a catalyst of the reaction between hydrogen and oxygen. This temporary inertness is attributable to a compact layer of adsorbed hydrogen.—J. S. G. T. W35 I.—Properties of Metals 5

*The Exact Measurement of the Specific Heats of Solid Substances at Higher Temperatures. XV I.— The Specific Heats of Metallic Thorium and o! Thorium Dioxide Between 20° and 1400° C. P. M. Jaeger and W . A. Veenstra (Proc,. K . Alcad. Wet. Amsterdam, 1934, 37, 327-332).— [In English.] The mean specific heat of thorium has been measured between room temperature and 1200° C. The atomic heat increases continuously from 8-235 at 300° C. to 11-785 a t 1200° C .— S . G . »Resistivity of Zinc Crystals. W . J. Poppy (Proc. lou-a Acad. Sci., 1932, 39, 2 1 1 ; G. Abs., 1934, 28, 6602).— In an attempt to settle the discrepancy between the resistivity measurements of Bridgman on the one hand and of Tyndall and Hoyem (./. Inst. Metals, 1931, 47, 645) on the other, single zinc crystals of 1 cm .2 cross-section and 10 cm. long were grown and measured. The results agree with those of Tyndall and Hoyem. Present indications are that certain anomalous crystals (i.e. not truly single) have abnormally low' resistivities and show great sensitivitv to slight strain.—S. G. ♦Influence of a Grain Boundary on the'Deformation of a Single Crystal of Zinc. Richard P. Miller (Metals Technology, 1934, (Oct.), A.I.M.M .E. Tech. Publ. No. 576, 1-9).— From tensile experiments at 180° 0. on single-crystal rods of zinc terminated by a transverse grain b o u n d a r y adjoining poly- crj stalline metal it is shown that the influence of the grain boundary is con fined to glide planes which it directly intersects and is composed of two distinct parts. In the first part, the glide layers are held motionless and the maximum extent of this influence is the distance p, such that d — p l [tan | (a -f- a')] + i>/[tan (90° — a)], where d is the original diameter of the rod, a is the original angle of the basal planes to the axis of the rod, and 7/ is their final angle thereto. In the second part, the glide layers have passed through a flexure and slid over one another to a small extent but not sufficient to form a uniform band ; the maximum extent of the influcnco of the grain boundary „ is <7, w here q — d cos a! . cos (a — a')/sin a. IJenco the grain- boundary influence varies regularly with the size of the boundary, the amount of deformation, and the ductility of the crystal.—A. R . P. »Influence of Chemically- and Mechanically-Formed Notches on Fatigue of Metals. Dunlap J. McAdam, Jr., and Robert W . Clync (,/. Res. Nat. Bur. Stand., 1034,13, 527-572 ; Research Paper No. 725).— An introductory section discusses the importance of stress concentration due to notches, as a cause of failure in service. Resistance of a notchcd specimen to fatigue and to impact may depend on entirely different properties. The influence of notches on fatigue is considered in the present paper. In section I I the cffcct of chemically- formcd notches is considered, attention being confined to the influence of p itt ing caused by stressless corrosion. Relationship between tensile strength and the % decrease in the fatigue lim it of steels and aluminium alloys is illustrated by composite curves, each of which is presumed to represent this relationship for a notch of fairly constant effective sharpness. The 3-dimensional relation ship between corrosion time, % damage, and the tensile strength is illustrated and discusscd. The general object of section I I I is to determine whether com posite curves of sim ilar form may be obtained by a studv of experimental data obtained by a number of investigators with mechanically-formed notches. The fact that such graphs have been obtained, each representing the influence of one form of notcli on one kind of metal, confirms the conclusion that a stressless corrosion graph of this type represents the influence of a notch of fairly con stant effective sharpness. Reasons arc discusscd for the deviation of individual results from the ideal composite curve for a mechanically formed notch. In section IV is considered the relationship between notch sensitivity (as measured by % damage) and other properties of metals. The properties considered are : hysteresis, ductilit}7-» and work-hardening capacity. Evidence is presented that scatter of individual results in a composite graph of the type used is due 6 Metallurgical Abstracts V ol. 2 largely to differences in tensile work-hardening capacity. Evidence is also presented that notch sensitivity, while depending somewhat on elastic hysteresis, depends largely on work-hardening capacity. The influence of notches in diminishing the advantage of superior strength is dealt with in section V. A bibliography of 33 selected references is appended.— S. G. *Elasticity. Arthur Cecil Vivian ( Inst. Civil Eng. Selected Eng. Papers, 1934, (150), 1-32).— An extension of Hooke’s law to enable more exact use to bo made of materials for constructional purposes is proposed. Elasticity is taken to be that property which determines that a material will completely or partially resume its original shape when the applied force is removed, and Hooke’s law is written: stress = {( + l ) — l } . f , w here S is elastic strain, S m ultimate elastic tensile strain, J a form factor denoting elastic character, and f ultimate tensile stress. When J = 2, this reduces to Hooke’s law.. Examples of the application of the method are given.— J. C. _ ’‘Elastic Failure of Thick Cylinders. Harris Booth (Inst. Civil Eng. Selected Eng. Papers, 1934,(138), 1-40).;—The conditions of failure are examined for thick cylinders subjected to internal pressure, to radial temperature gradient, and (for cylinders closed at one end) to both internal and external pressures. It is Shown that no stresses aro induced in a thick cylinder subjected to a uniform longitudinal temperature gradient.— J. C. C. *On Hardening Phenomena in Pressed Metal Bodies. W . Trzebiatowski (Z. physikal. Chem., 1934, [B ], 24, 75-S6).— Bodies made by pressing very finely- divided gold and copper powders at pressures up to 30,000 atm. show consider able hardening effects, characterized by a broadening of the interference lines in the rontgenogram and a high degree of dispersion; no true texture is, how ever, produced. The hardness values obtained are much higher than those observed on either metal after severe working; this is attributed to the very fine crystal structure of the pressed bodies. The fall in hardness which occurs on annealing is produced by recovery and recrystallization phenomena.— K . S. *0n the Problem o£ the Electrical Conductivity of Synthetic Metal Bodies. ■W. Trzebiatowski (Z. physikal. Chem., 1934, [B ], 24, 87-97).— Compared with massive metals, pressed bodies made from metal powders show peculiar anomalies in the electrical conductivity. For pressed bodies of gold and copper the temperature coeff. of resistance is positive up to 100° C., negative between 100° and 300° C., and positive again at higher temperatures; this behaviour is attributed to gas films and to recrystallization phenomena, which have been confirmed by dilatometric measurements.— K . S. ^Researches on the Thermal Conductivity of Metals (W ire or Tape) at High Temperatures. M. Conard ( Aerotechnique (Suppt. to Aeronautique), 1934, 12, 26-27).—A method is briefly described and illustrated in which the wire is mounted in an evacuated tube between 2 tapes so that it can be heated by a current passed either through it or through the supporting tapes alone. —j . C. C. A Classical Model of a Ferromagnetic Material and Its Subsequent Quantiza tion in the Region of Low Temperatures. G. Heller and H. A. Kramers (Proc. K. Akad. Wet. Amsterdam, 1934, 37, 378-385).— [In German.] The classical model of a ferromagnetic material is defined; only when conceived in terms of a space lattice does the model indicate the occurrence of ferro magnetic saturation. Quantization of the model produces the formula: of Bloch and Moller exactly.—J. S. G. T. *Drift of Magnetic Permeability at Low Inductions after Magnetization. Raymond L. Sanford (./. Res. Nat. Bar. Stand., 1934, 13, 371-376; Research Paper No. 714).— The magnetic perm eability of ferromagnetic materials at low values of induction depends on the time which elapses between demagnetization and testing. The change may be of the order of 10 or 12%. In order to 1935 II.—Properties of Alloys7

obtain consistent and reproducible results in testing at low inductions, a period of from 18 to 24 hrs. should elapse after demagnetization before the tost is made.— S. G.

II.—PROPERTIES OF ALLOYS

‘ Correlation o£ Equilibrium Conditions in Binary Aluminium Alloys. W . L. Fink and H. R. Freclie (Metals Technology, 1934, (Oct.), A.l.M.M.E. Tech. PM. No. 580, 1-14).—The results obtained previously for the equilibria at the aluminium end of binary systems of that metal with other metals of high purity have been analyzed mathematically, and the solid solubility and hypereutectic liquidus curves are shown to be in agreement w ith those deduced from the thermodynamic equation log, x' — — LIRT -+- C, w here x' is th e mol. fraction of alloying element, L the molal heat of solution, R th e gas constant, T the absolute temperature, and C the integration constant. The systems with magnesium and with magnesium silieide, however, do not obey this rule. A straight line is obtained by joining the points obtained by plotting the logarithm of the eutectic lowering against the logarithm of the atomic-% of solute in solution at the eutectic temperature, the point for manganese being the only one not near this line. The reciprocal of the slope of the solubility lino for all binary systems is a linear function of the logarithm of the concentration of the solute, and therefore the slope of the solid solubility curve for any binary aluminium system can bo determined from any point on it by means of the expression : jl _ r log x ,______T ______. r> p S 0 0003T - 1 0-417 - 0-0001252'' ' ' ' ♦Effect o£ Quenching Strains on Lattice Parameter and Hardness Values of High-Purity Aluminium-Copper Alloys. Arthur Phillips and R. 51. Brick (Metals Technology, 1934, (Sept.), A.l.M .M .E. Tech. Publ. N o . 563, 1-19).-—- Coppcr-aluminium alloys made from very pure metals may show an abnorm ally large lattice parameter when quenched from the solid solution range due to strain resulting from quenching stresses. The parameter is greater the more severe the quench, the greater the diameter of the specimen up to 0-5 in., and the higher the copper content of the solid solution. The maxi mum age-hardening capacity at 20° C. is shown by drastically qucnched alloys with 5-4% coppcr; during ageing there is first a slight decrease in the lattice parameter, but the constant value ultim ately obtained is sufficiently high to indicate that partial precipitation is the cause of the observed harden ing. The initial decrease in parameter is attributed to gradual relief of stress. When the quenching is only just sufficient to retain the metastable solid solution without producing severe quenching strains the alloys do not exhibit any hardening after 30 days at 20° C., which suggests that the presence of these strains is essential for hardening to occur at room temperature. Pre cipitation of CuAl, at 275°-325° C. is more rapid the greater the quenching strain and the greater the degree of supersaturation; when the alloys are again quenched after precipitation is complete the parameter of the residual solution is greater than that of pure aluminium, but when they are air-cooled or quenched in boiling water a normal reaction curve is obtained. Reheating at the precipitation temperature followed by a drastic quench again pro duces a distended lattice, which cannot, therefore, be due to strain induced by the precipitation process. Maximum hardness after ageing at 275°- 300° C. is obtained when precipitation is practically complete.—A . R . P. ‘ Ternary Diagram of the Aluminium-Copper-Silicon System. Kanji .Matsu yam a (Kinzoku no Kenkyu, 1934, 11, (10), 461—190).— [In Japanese.] The diagrams of the binary systems aiuminium-copper, copper-silicon, and 8 Metallurgical Abstracts V ol. 2

alunnnniin-silicon have been carefully redetermined by means of thermal analysis, electrical resistance determinations, and microscopic examination. alummuim-coppcr-sihcon system was then investigated from melt to room temperature, and the ternary diagram of the whole system was con- stmcted.— b. G. *The Influence of Pickling on the Fatigue-Strength oï Duralumin. H. Sutton and W . J. Taylor (J. Inst. Metals, 1934, 55, 149-158; discussion, 158-164).— ? „ . eot 0.n.t,1?0 fatlg',e hunt of Duralumin in Wohler-type fatigue tests of the iol lowing piekhng treatments has been determined: (.4 ) 2Î- minutes in 10% caustic soda solution at 60°-70° C., rinse, 1 minute in 10% nitric acid containing V/ 9 of sulphuric acid; (B) 2 minutes in a solution containing 10% each of nitric and hydrofluoric acids ; (C) 3 minutes in a 4 : 1 mixture of 10% sulphuric and hydrofluoric acids, rinse, 1 minute in 50% nitric aeid. The/percentage decreases in the fatigue lim its observed were (^1) 31%, (B) 1 5 % , (C ) 6 % * im- uîî)1' c?n/aft nvthe Pickling treatment reduced these decreases y ) 10 o and (C) 3-8/a. Treatment (/l) produces a rough serrated surface, and treatment (C) reveals the macrostructure of the alloy and is, therefore, suitable for the examination of the alloy for defects in manufacture. Removal oi the pickled surface layer by machining completely eliminates the effects of piekhng on the fatigue lim it, indicating, as suggested by D. Hanson and by . U. biatcr in the discussion, that pickling produces a weakening of the surface layer cither by gas adsorption or by setting up internal stresses, or that inter- crystallme penetration of the pickle occurs and induces eorrosion-fatiguo (suggested by It. Sehgman). K. Wood stated that for commercial work a mixture of sodium fluoride and sulphuric acid was preferable to hydrofluoric acid for piekhng light alloys.— A. R . P. .. *St.U(îif1s on Decomposition of Supersaturated [Solid Solutions] in Light fri , ^s' Schmid and G. Siebel (Melallwirtschaft, 1934, 13, 765-768). ~7 “ eî 1° ”’P?s*t:i01} quenched alloys of magnesium with zinc and alumi nium at 218 C., and of aluminium with magnesium at 218° and 155° C. has been followed by X-rays. The precipitation takes place the more rapidly the greater the degree of supersaturation of the solid solution and more slowly with single crystals than with polycrystalline aggregates. In zinc-magnesium alloys and m single crystals of aluminium alloys the composition of the solid solution decreases steadily from the beginning to the end of the reaction. In polycrystaihne alummium-magnesium and magnésium-aluminium alloys the initial and final concentrations are close together. The changes in the meehani- cal properties follow closely the progress of precipitation.— y. G. Electrometallurgical Research and Its Relation to the Grand Coulee Pow °r Development (Some Facts on the Ultra-Light Structural Alloys of Magnesium and Alummium). A. E. Drucker (State Coll. Washington, Met. lies. Bur Information Cm. N o . 8, 1934, 0 p p .).— S . G . Silumin-Gamma. J. Domauf ( Alum inium , 1934, 17, 26-31).—See also Met. Abs., 1934,1,414. Silumin-y is the eutectic silicon-aluminium alloy con taining small additions of magnesium and manganese and modified by the usual sodium treatment just prior to pouring. After quenching from just below the solidus the alloy can be hardened by an ageing treatment, Sand-castings have , ar CSS,0f 8?~100,'!l y iol.fl-P°iut °f 18-25 kg./mm.2, a tensile strength , kg-,mm. , and an elongation of 4-0-5% ; the corresponding values for ch.ll-castmgsare 85-110, 20-28, 26-32, and 1-5-0-5, and for die-castmgs 110- , J , . ’ a n « 1-1-5. T h e a llo y finds ex tensive use in th e m an u factu re of housings for aero- and automobile engines and sim ilar intricate castings where lightness, strength, and a high yield-point are desirable.— A. R, P. ° Change of Mechanical Properties of Copper-Rich Alloys Due to the Grain- Refinement hy the Peritectic Reaction. Ju-n Asato (Iiinzoku no Kenkyu, J3 4 , 11, (8), 36o 376). [In Japanese.] The mechanical properties such as 1935 II.—Properties of Alloys 9

Brinell hardness, tensile strength, elongation, yield-point, and lim it of pro portionality of some copper-rich alloys were determined to show how these properties are modified by grain-refinement causcd by peritectic reaction. — S . G . ♦Crystal Densities of Industrial Brasses from X-Ray Data. E . A. Owen and Llewelyn Pickup (J. Inst. Metals, 1934, 55, 215-222; discussion, 223-228).— The density of homogeneous alloys of copper and zinc in the a, a 4- jS, and p regions have been calculated from the lattice parameter as measured by X-rays, and the results obtained are shown to be superior to those obtained by weighing in air and water, since they are unaffected by porosity, cold-work, and grain- size, as well as by heat-treatment in the case of pure a- or pure jS-alloys. The density of a-alloys is not a linear function of the composition, but that of /3- and that of (a + $)-alIoys can be taken as linear to a high degree of accuracy; at the phase boundaries, however, discontinuities in this relation occur. The values obtained are tabulated and the effect of quenching temperature on the density of duplex alloys is shown graphically.— A. R . P. *Gold-Cliromium Resistance Alloys. James L. Thomas (,/. Res. Nat. Bur. Stand., 1934, 13, 681-688; Research, Paper No. 737).— The addition of from 1-6 to 2-4% or more of chromium to gold produces alloys having very small temperature coeff. of elcctrical resistance. In particular, 2-1% chromium in gold gives an alloy whose resistance has been made independent of temperature, to a few parts in 10 million, over at least the interval 20°-30° C. These alloys are also exceptionally stable in resistance. They have, however, a thermo electric power against coppcr which is 3 or 4 times as large as that of Manganin. rIh e preparation and heat-treatmcnt of some of these alloys are described. — S . G . *Age-Hardening of Lead Alloys. H. Hariba (Sumitomo-Densen Iho, 1934, 1, (2 ), 49-57; C. Abs., 1934, 28, 7229).—Lead-calcium alloys containing 0 -02% calcium undergo age-hardening; the maximum hardening effect is at 0-1% calcium. The rate of change in hardne$8 and in elcctrical resistanco of lead containing calcium on ageing varies according to the calcium content; the volume of lead containing calcium contracts a little on age-hardening. In lead containing antimony expansion is shown.— S. G. Bearing Metals in Use on the Railways of the U.S.A. [Satco Metall. [Fr.] W it te (Organ Fortschr. Eisenbahnu'esens, 1934, 89, 400-402).— See Met. Abs., 1934, 1, 416.— P. M. C. R. ♦Solubility of Carbon in Iron-Manganese-Silicon Alloys. C. H. Herty, Jr., and M. B . Royer (U.S. Bur. Mines, Rept. Invest. N o . 3230, 1934, 22 p p .).— T he solubility of carbon in iron-manganese alloys up to 75% manganese, in iron— silicon alloys up to 50% silicon, and in iron-manganese-silicon alloys with a 4-10 : 1 manganese-silicon ratio has been determined at 1300°-1706° C-, and the results are shown in tables and graphs. Manganese, having a carbide- stabilizing action, increases the solubility of carbon, whereas silicon, having a graphitizing action, has the opposite effect, only 0 -1% carbon being retained in solution in the 50 : 50 silicon—iron alloy. A 3-page table is given showing the carbon solubility at 1300°, 1500°, and 1700° C. in alloys containing 0-80% manganese and 0-20% silicon in 5 % steps. The bearing of the results on the composition and use of these alloys as steel deoxidizers is discussed.—A . R . P . *On Anomalous Properties of New Magnetic Materials [Copper-Nickel- Iron Alloys]. Martin Kersten (Fiss. Verdff. Siemcns-Konzem, 1934, 13, (3 ), 1-9).— The magnetic properties of 45 : 55 nickel-iron alloys with additions of 0 ,3 ,6,9,12, and 15% copper have been determined after annealing, in the form of 3 mm. wires, at 900° O. lor 2 hrs., slowly cooling, and rolling to strips 0-08 mm. thick without further annealing. Without addition of coppcr the alloys g ive a norm al m agnetization cu rve and a rem anence of about 5 0 % of th e sa tu ra tion magnetization, whereas the 9 % copper alloy gives a linear magnetization 10 Metallurgical Abstracts V o l . 2

curve and a remancnce of only 6 % , which, however, is increased almost to 100% if the alloy is subjected to a tensile stress of about 60 kg./mm.2 in the direction of the magnetizing field. Analysis of the magnetic stresses in these alloys indicates that the stress rises sharply with increasing copper content; anneal ing tests indicate that this increase in stress is due to precipitations sim ilar to those which occur in prccipitation-hardcning. Very high rcmanencc values approaching closely to the saturation magnetization can be obtained in the 12 and 15% copper alloys by annealing the hard-rolled strip at 600° C.— A. P. *Torsional Moduli Variations of Spring Materials with Temperature [Konel]. Joseph W . Ludewig (Trans. Amer. Soc. Metals, 1934, 22, 833-860).— T he behaviour of springs at elevated temperatures involves knowledge of tho values of tho modulus of rigidity and proportional lim its of materials at such temperatures. A summary of previous work on the variation of the modulus of rigidity with temperature, and on the behaviour of springs at elevated temperatures is given. The torsional moduli of a number of materials are plotted against temperature, and in accordance with the procedure outlined in the paper the following results were obtained: (1) Materials maintaining highest absolute value of the modulus of rigidity up to 450° F. (232° C.) aro high-speed steel, stainless (cutlery) steel, and Konel. (2) Material maintain ing highest absolute value of the modulus of rigidity up to 985° F. (529° O.) is high-speed steel. (3) M aterial showing lowest absolute value of the modulus of rigidity at all temperatures is carbon spring steel. (4) The remaining m aterial tested was Silcrome steel. A discussion of the factors found in spring formulae and behaviour under temperature variations is included.— S. G. Metallurgical Applications of Silicon. J. W . Donaldson (Metallurgia, 1934, 11, 20-22).—The uses of silicon as an alloying metal are considered and its applications in steels, cast irons, corrosion-resisting alloys, and non- ferrous alloys are discussed. The non-ferrous alloys referred to aro tho silicon-aluminium alloys of the Duralumin, R .R. alloys, Alpax (Silumin), and hypoeuteotic aluminium-silicon alloy types; the coppcr-silicon alloys, both silicon-bronzes and silicon-brasses, and nickel-silicon alloys.— J. W . D. Recent Development in Main and Connecting-Rod Bearings. Stanwood W . Sp arro w (Soc. Automotive Eng. Preprint, 1934, June, 7 pp.; and (summary) Automotive hid., 1934, 70, 772—773).—Evidence is given which indicates that flexing and faulty bonding are not tho primary causes of cracking in main and connecting-rod bearings of automobile engines, but that cracking is produced by a tangential force between the journal and the bearing at spots where the oil-film is inadequate to prevent metal to metal contact. A minimum thickness of 0-03 in. of Babbitt metal is recommended, and under cutting has been found to extend the life of the bearing since it prevents the cracks from spreading and permitting the Babbitt metal to escape. Copper- lead bearings (lead 45, nickel 2, copper 53% ) crush as readily as Babbitt metals, and metal to metal contact produces a lead film which has sufficient lubricating value to prevent seizing; failure of copper-lead bearings occurs by disinte gration of parts where it is difficult to m aintain an oil-film or when an unsuitable 011 is used. Lubrication of bearings and the effects of “ dirt ” are briefly discussed.—A. R. P. fThe Improvement o£ W hite Bearing Metals for Severe Service : Some General Considerations. D. J. Macnaughtan (J. Insl. Metals, 1934, 55, 33-47).— Development in the internal combustion engine is imposing increasingly severe conditions on bearings. Consideration is given to the theoretical functions of an ideal white metal, and the manner in which the stresses produced in service tend to cause failure by cracking. Since the normal action of the stresses is compressive, special attention is given to the tension stresses which are shown to lower the fatigue range of the metal and to open up incipient cracks. Based on this analysis the mechanism of crack formation is discussed. The following 1935 II.—Properties of Alloys 11 directions in which improvement in service behaviour may be secured arc considered: (1) diminishing the intensity of the stresses in the metal by modi fications in (a) certain features in design; ( b) the material used for the liner; (2) increasing tho fatigue-resisting properties of the white bearing metal, in respect to which results obtained in preliminary investigations of the fatigue properties of high tin-antimony-copper alloys with and without addition of a further element aro given.—D. J. M. ♦The Behaviour of W hite Bearing Metals when Subjected to Various Deforma tion Tests. Introduction. ---- (J • Inst. Metals, 1934, 55, 49-50).— The scope of tho investigation and compositions of the materials used arc given. The results of the research aro described in 3 parts; for abstracts see below.-—S. G. ♦The Behaviour of W hite Bearing Metals when Subjected to Various Deforma tion Tests. Part I.—Indentation Tests. A. S. Kenneford and Hugh O’Neill (J. Inst. Metals, 1934, 55, 51-69).—The effect of viscous flow, ageing, and pro longed heating on the resistance to indentation of tin- and lead-base bearing metals has been investigated. Flow tests with a 120° steel cone at 19° and 96° C. show that Babbitt metal containing 1% cadmium or 2% nickel, or a lead-alkali bearing metal, give better indentation results than a plain Babbitt alloy. The hardness of the different metallographic constituents of bearing metals and their softening on heating to 100° C. were measured by scratch and micro-indentation tests. Tho matrices lose 40-45%, and the cuboids 20% of th e ir hardness, b u t th e cuboids in a B a b b itt rem ain som ew hat hard er th a n those in a lead-base alloy. Two new simple tests aro suggested. In the first a lubricated 60° conical casting of the alloy is flattened under 100 kg. load for 30 seconds, and the Mallock hardness number determined. B y increasing the duration of loading a flow index may be measured on lines similar to “ H ar greaves’ analysis.” Then, by compressing until cracks appear on the extruded edge, the ductility of the specimen and its cracking stress may bo measured. A t room temperatures tho lead-base alloys show relatively low ductility, and this agrees with their low work-hardening capacity as determined by specially conducted ball tests and repeated impact tests with tho scleroscope. The second method employs an instrument similar to tho Herbert pendulum, and measures the damping effect. It may not only be used to give rapid indications of hardness at different temperatures, but is also sensitive to tho effcct of dif ferent lubricants.—A. S. K . ♦An X-Ray Examination of the Phases in Babbitt Metal, and of the Age- Hardening of Cast Lead-Alkali Alloy. G. S. Famham {J. Inst. Metals, 1934, 55, 69-70).— Appendix to a paper by Kenneford and O’Neill (see preceding abstract). An alloy of the composition SnSb after annealing for 1 week has a structure of tho NaCl type with a = 4-099 A . T h e presence of th is com pound in Babbitt metal has been confirmed by X-ray tests. The needle constituent of Babbitt metal has been isolated by liquation and its composition proved to be CuSn or the -q-phase of the copper-tin system. X-ray examination of a sodium-calcium-lead bearing alloy in the newly cast and in the aged condition shows that the cast alloy consists of two phases, one of which changes w ith age ing whilst the other does not; the latter is face-centred cubic with a — 4-889 A ., and is probably CaPbs. The former is a supersaturated solution of sodium in lead which deposits a sodium-rich phase (possibly Na2Pb j) on ageing.—A. R . P. ♦The Behaviour of W hite Bearing Metals when Subjected to Various Deforma tion Tests. Part H.— Tensile Tests. R. Arrowsmith (J. Inst. Metals, 1934, 55, 71-76).— The tensile properties of white metal specimens, prepared by gravity die-casting and without any machining, have been determined at room tempera ture on a Hounsfield “ Tcnsometer.” Various casting conditions were examined for each alloy. Babbitt metal with additions of cadmium gave the highest values of lim it of proportionality and ultimate stress. The greatest ductility was obtained from an alloy containing 89% of tin.—R . A. Metallurgical Abstracts V o l. 2

*T^le ® eha£I0l'r ° £ W hite Bearing Metals when Subjected to Various Deforma- H - Gree'"™ od (J- I n J V J ü Z m k inà of white mô a l! v,C form of the Stanton impact tester suitable for the 'test- «n?t=„ r 1“c.till.s b> Pounding at different temperatures is described Rc- W nW C-' ni specimens are given, and the unsuitability of this type of test-piece is shown The use of bcaring-shaped specimens with a cylindrical allovs and aî^ rl r E f,Slllts arP rc«»'ded i« 8 different white-metil bearing 100° and 150®^ A Rabunder various conditions, and tested at 18% ,A Babbitt metal with an addition of cadmium cave the greatest resistance to pounding.—H. G. ** _ Joint Discussion [on the Improvement o£ White Bearin'* Metals for W o ™ Service . — (./ Inst. Metals, 1934,55,88-113). L e 6 S c < S b s tr^ te -

i . mctal13 m a cold-worked condition in which the rate of creen under load is much greater than when the metal was cast • figures and curved

t 6 ^ ^ ^ T O t^ v S ’ehat ?Ur' iHC had f0lmd that additi0n of cadmium cub“ ttheed'esPofw • ,CraC^ "S 4urmg cold-rollinK by suppressing the the failure of b ifl, ^ • crackm/ commenced. II. Sutton considered that addition ?f O-ô^ nickeMo^?! fotigue-cracking and stated that respect H L 1 m % * a!1,°yS lmProv« their behaviour in this cadmium-base :b f w 3 that annealing improves the stress conditions in c annum case bearing metals ; those alloys can now be rcadilv cast on to K th s C d lS m ill^ glV° Valf S °/ °'5 tons/in-2 in the shear of bond test and o withstand lo million reversals of stress at a stress of 18-9 tons /in 2 on ihn ihere fa fS te d l?1 "w * K Baile^ a "»tlfcmatical p^oofthat V R do.“ I}lte relation l>etween mdenter flow tests and tensile creep tests H. A . Bassett described a case of failure of a 50: 38 : 10 • 2 tin-lead-antimonv ' » r d a!,a tearin« metal in the axles of railway wïons Ot/er speakers pu, forward various theories to account for the cràèkinc of Babbitt m etal bearings subject to high duty, and Macnavtqbian, Kenneford O’Neill and

for S ÎJé ïÈ S S ^ 8 P “ 1 ,™ td ConnT6 ?f°w rtieS V mr S “ * SmaU Amounts of SUver, Iron, Nickel or 115-131. S : r1 ?vsternst({- ** and conner-tin cnuililir' ' r tm-rich ends of the silver-tin, nickel-tin, investigated. W ith the firrt, the at a te!nTiera v,i . wl?;^ . 2213 C‘: mth tho second, at 0-18% nickel

7 ° . I Uün

ï h c m e t h ° d copiw and rdc’kel P T it infl d,fficulttles “ re met with in the case of silver, iron, Q+tvvii Ifi r r • v I-* influence of additions of these metals on the tensile a llo v f fin 18 cusscd- A great- increase produced b v quenching s ilv e r- tin quenchh ™ I n f no effect LH r te,nP?rat,Ire> " hil^ with the other 3 alloys E t u n to thi t Additions of iron above 0-4% are without effect, fomd“ 8 M ?k e îu S o ^ W n^e, an mcr?ase of 40% in thc tensile strength is additions have 7o b O ^ T o S ^ - ^ strength after all heat-treatments investigated. s K re Ë e ^ th e & o fff but does not prevent gram-growth at high temperatures. The addition of iron 1935 II.—Properties of Alloys 13 above 0-05% or of nickel above 0-06% prevents such grain-growth, although below these compositions germination takes places. 0-35% and more of copper prevents rccrystallization of cold-rolled tin at room temperature, but annealing at temperatures from 110° C. upwards produces larger grains than in alloys of slightly lower copper content.— D. H. *Studies of Phase Changes During Ageing of Zinc-Alloy Die-Castings. I.— Eutectoidal Decomposition oi Beta Aluminium-Zinc Phase and Its Relation to Dimensional Changes in Die-Castings. 51. L. Fuller and 11. L. Wilcox ( Metals Tccknologi/, 1934, (Sept.), A.l.M.M.E. Tech. Publ. No. 572, 1-17; and (abstract), Iron Age, 1934, 134, (15), 30-31).— On ageing tho quenched /3-aluminium-zinc alloy at 0° C. decomposition into a + y is complete in 3 J minutes and the temperature rises to 84° C.; the corresponding figures for ageing at 20° C. are l| minutes and 108° C., and at 50° C. 1J minutes and 133° C. After quenching from 350°-375° C. in water at 20° C., tho 2 0 % aluminium-zinc alloy decomposes into a + y in 1J minutes with an increase in temperature to 99-3° C.; w ith 1% copper the time required is 58 minutes and the temperature reaches 78° C., an d w ith 0 -1% magnesium decomposition starts in 3 days at 20° C. and is completed in about 35-40 days at 20° C., but at 100° C. decomposition is complete in 1 day. The change in length on ageing the /3-alloy with 0-1% magnesium is 0-29%. The ¿¡-phase in com mercial zinc-base die-casting alloys with 0 '3% copper and 0-04% magnesium is completely decomposed in less than 1 day; this may be due to the slower rate of cooling or to the acceleration of tho decomposition by the large amount of a present. The shrinkage of the commercial alloys cannot bo due entirely to decomposition of /J; it is highly probable that other phase changes occur which also cause shrinkage. The heat of decomposition of pure /9 is 7-39 grm.-cal./grm. for the 20% aluminium alloy and the decomposition temper ature is about 277° C.—A. R . P. *An Analogy Between Plastic Deformation and Certain Cooling Rates in Causing “ Premature ” Precipitation in Supersaturated Solid Solutions. The Incubation Period,—I. J. L. Burns (Trans. Amer. Soc. Metals, 1934, 22, 728-736).— Reports a study of the agc-hardcning of Duralumin. The effect of rate of cooling on age-hardening and on the incubation period was inves tigated. It was found that if a supersaturated solid solution is plastically deformed immediately after the quenching operation, ageing begins earlier, there is no incubation period, and the final hardening due to precipitation is less. If a particular rate of cooling is used in quenching from the temperature of maximum solubility, ageing begins earlier, there is no incubation period, and tho final hardening w ill usually bo less. From this it is concluded that plastic deformation does not increase tho rate of hardening, but merely causes the hardening to start earlier. Precipitation may take place during a cold- working operation or during quenching. When an incubation period ensues after quenching, it is suggested that the solute has been retained in its entirety during the quenching operation. The hardness of quenched steel is prob ably due to the same cause as that of oil- or air-cooled Duralumin, namely, ageing during the cooling operation. This analogy, if true, obviates any sharp lino of demarcation between ferrous and non-ferrous alloys.— S. G. f*The Present Status o£ Age-Hardening. Richards H. Harrington (Trans. Amer. Soc. Metals, 1934, 22, 505-524; discussion, 525-531).—The theoretical aspects of age-hardening are briefly outlined for 3 general types: (a) sim ple precipitation age-hardening, (b) simple lattice-strain age-hardening, and (c) complications due to allotropy. The practical significance of age-hardening alloys at elevated temperatures in relation to tensile strength and for spring properties is discussed, and an interesting plan is given for exploring an unknown system. Cobalt added to age-hardening alloys acts in general further to increase the agc-hardness and often to increase the temperaturc-of- 14 Metallurgical Abstracts V ol. 2

maximum age-hardness. It may act generally in 2 ways : (1) as a “ desol vent,” reducing the solubility of the precipitating constituent, at least part of the cobalt going into solid solution in the solvent lattice; ( 2) almost all of the cobalt (if correctly added) enters into a ternary constituent with the precipitating agent, thus usually decreasing the solubility of this precipitant. — S G fAge-Hardening Alloys. ---- (Metallurgist (Suppt. to Engineer), 1934, 10. 149-160).—A brief review of the subject and a summary of a paper by R . H. Harrington (preceding abstract).—R . G. ♦Method of Radiation-Calorimetry for Determining the Thermal Conductivity of Metal Bars. P. Vernotte (Akrotechnique (Suppt. to Aironauliqm), 1934, 12, 25).-—A brief, illustrated note of a method for determining thermal con ductivities by observing the power dissipated by a small electric furnace when a bar of the metal is inserted in the furnace in different positions. The conductivity of a brass bar 50 mm. long and 4 mm. in diameter was found to be 0-29 cal./deg./cm ./sec.— J . C . C. tAlloys [The Hume-Rothery Rules]. Anon. (Metallurgist (Suppt. to itiiujvtieer), 1934, 10, 174—176).— A brief review of the subjects of codification of data relating to equilibria in alloy systems and the formulation of rules on which to base reliable predictions. A paper by Hume-Rothery, Mabbott, an d E v a n s (Met. Abs., 1934, 1, 296) is critically discussed. It is pointed out that the false valency factor or “ liquidus factor” of Hume-Rothery and his collaborators, calculated from data on copper and silver alloys, is not a whole number as claimed. In consequence, theoretical arguments based on this premise must be iegarded as invalid, and the scientific basis of liquidus curves as not yet known. In the prediction of the freezing points of ternary alloys, the authors referred to have been singularly succcssful, and though their results are regarded as empirical, the line of investigation followed should provide a measure by which experimental results may be compared and codified.—R . G. Metallic Alloys. Anon. (Werkstatt u. Betrieb, 1934, 67, 353-354).— A summary, giving the average composition of a number of alloys, classified as : copper alloys (not bronzes), bronzes, aluminium alloys, nickel alloys, zinc a oys, tin alloys, lead alloys, white (bearing) metals, fusible alloys, light alloys, amalgams, silver and gold alloys.— P. M. C. R . A Metallurgical Survey of Engineering Material. Josiah W . Jones (Met. 1ml. (Land.), 1934, 45, 537-541).— An abstract of the “ non-ferrous ” portion of a paper dealing with both ferrous and non-ferrous metals, read before the .Birmingham Branch of the Institution of Production Engineers. The strength to weight ratio of heavy and light alloys, the “ shapeability ” of alloys, the theory of hardening of Duralumin, the hardening of copper alloys, particularly of Kunial, permanence as regards the fracture of defective material and the weakening of section due to corrosion, quenching non-ferrous alloys, and the properties of heat-resisting alloys are discussed.— J. H . W .

III.—STRUCTURE (Metallography; Maerography; Crystal Structure.)

Etching Polish for the Preparation of MetaUographic Samples. N. Zarubin, M. buitm, and N. Golikov (Zavodskaya Lab., 1933, (7), 49-53 ; C. ^ 65., 1934, [In Russian.] Osmond’s method of polishing samples for metallo graphy analysis with an etching polish was used successfully not only for metals such as tungsten, molybdenum, tantalum, vanadium, &c., but for non- ferrous metals and alloys such as copper, aluminium, nickel, tin, lead, bronze, &c. Ib references are given.—S. G. 1935 III. —Structure’ 15

*The Deformation Lines in Alpha Brass. Carl H. Samans (■/. Inst. Metals, 1934, 55, 209-213).— A microscopic study of 70 : 30 brass single crystals of 2 different orientations which had been reduced 50% in thickness by cold- rolling revealed the presence of many of the so-called “ lines of deformation.” X-ray determinations in the rolling plane showed conclusively that the mark ings were mechanical twins parallel to octahedral planes.— C. H. S. ♦An X-Ray Study of Orientation Changes in Cold-Rolled Single Crystals of Alpha Brass. Carl H. Samans (Metals Technology, 1934, (Oct.), A.I.M.M.E. Tech. Publ. No. 579, 1-15).—The lattice rotations found in rolled single crystals of 70 : 30 brass of 5 different initial orientations have been successfully explained on the basis of plane parallelepipedal compression representation of the forces of rolling. The active slip plane can be determined both from the maximum shear-strcss law and the manner of rotation. One of 3 slip systems m il be active according to the rolling plane and direction of rolling chosen, and the lattice rotations produced by deformation -will be a variation of one of the following ideal positions of rolling plane and rolling direction: (110), [112]; (110), [001]; (110), [O il]; (001), [010]; (112), [111].—A. R. P. ♦Crystal Orientation in Drawn Brass Cups. L. Herrmann and G. Sachs (Metallwirtsckift, 1934, 13, 745-752).— The degree of deformation and the orientation of the crystals at various parts of a drawn 03 : 37 brass cup have been investigated; the observed textures correspond with those observed in simple deformation processes, thus where pure compression occurs the corre sponding fibre texture is found with [ 110] as the axis of the films, and where elongation occurs a double-fibre texture is formed with the [ 111] and [ 200] orientation. In the other zones the complex textures formed correspond with the complex nature of the deformation; no texture equivalent to the rolling texture can be found in any part of the drawn cup.— v. G. ♦Crystal Re-Orientation on Heating Drawn Copper Wires. G. S. Farnham and Hugh O’Neill (J. Inst. Metals, 1934, 55, 201-208).— The behaviour of a silver-free coppcr wire reduced 59% by cold-drawing, has been compared after “ low-temperature treatm ent1! (L.T.T.) with that of 2 silver-bearing wires reduced 59 and 49%, respectively. L.T.T. hardening occurs in the first, but not in the second of these. X-ray spectroscopy makes it evident that preferred orientation is less developed in these silver-bearing wires. The general effect of L.T .T . at 130° C. is to reduce the amount of [111] preferment, but to cause an increase of [100] preferment. This change-over probably causes “ orientation hardening.” In these silver-bearing wires, however, the change is only rela tively small in extent, and this appears to explain the differences as regards L.T.T. hardening.— G. S. F. ♦An X-Ray Study of the Effect of Heat on the Structure of Sputtered Films of Gold. S. Rama Swamy (Proe. Phys. Soc. (Lond.), 1934, 46, 739-746).— Sputtered films of gold of different thicknesses were heated to various tempera tures up to 800° C. and their Dobye-Scherrer X-ray photographs determined at each stage of heating. The photographs indicated that on being heated the gold crystals in the films oriented themselves so that their 111 planes were parallel to the surface of deposition. The crystals grew when the films were heated. The degree of orientation and size of crystals depended on the thickness of film and on the temperature to which the film was heated.— J. T. ♦The Crystal Structure of the Intermediate Phase Au 2Pb. Harald Perlitz (Keemiateated, 1934, 2, (1 ), 11—16; Chfin. Zentr., 1934, 105, I I , 2656—2657).— The phase is cubic face-centred, a — 7-910 A ., w ith 24 atom s in th e u n it ce ll. The structure of the coll is analogous to that of MgCu, and K B i 2.— A . R . P . ♦The Growth of Metal Crystals in Metal Vapours. H I.[—Magnesium]. M. Straumanis (Z. physikal. Chem., 1934, [B ], 26,246-254).— To test Kosscl and Stranski’s theory of the homopolar growth of crystals, experiments have been made on the sublimation of magnesium in hydrogen at 0 001-300 mm. pressure 16 Metallurgical Abstracts Vol. 2

at temperatures below the melting point. Earlier work on cadmium under similar conditions is reviewed and amplified by new measurements of the crystals produced. The magnesium crystals are bounded by [0001], [1010], and [ 1011] planes; unlike zinc and cadmium crystals they do not grow in layers. The results are in accordance with theory if the calculations take into consideration only the nearest neighbours to a particular lattice point.— K . S. Crystal Structure of Tl7Sb2. F. R . Morral and A. Westgren (Svensk Kem. Tids., 1934, 46, 153-156; O. Abs., 1034, 28, 7102).—The diagrams of ideal and real molecular structure of Tl,Sb, are given, together with a table of constants.— S. G. »Crystal Structure of Thallium-Bismuth Alloys. Arno Olander (Z. Krist., 1934, 89, 89-92; C. Abs., 1934, 28, 7229).— [In German.] Two intermediate phases in the thallium-bismuth system were observed. W ith a composition between 73 and 96% thallium, the structure is face-centred cubic, with a v a ry in g from 4-928 to 4-842 A . B e tw e e n 34 and 4 6 % th a lliu m the stru c tu re is hexagonal, with a = 5-670-5-642 A . and c = 3-369-3-375 A — S . G . *The Mechanism of the Transformation o£ Body-Centred Cubic Zirconium into the Hexagonal Modification. W . G. Burgers { Metallwirtschaft, 1934, 13, 785-786).—Recent work has shown that structural changes take place by an ordered translation in the lattice. The transition from a body-centred cubic to a close-packed hexagonal lattice can therefore occur only by the formation of a face-centred cubic lattice as an intermediate phase; this appears to be the case with calcium (cf. Graf, Met. Abs., 1934, 1, 496).— v. G. *On the Influence of the Method of W orking on the Orientation of Crystals of Wire. G. von Vargha and G. Wassermann (Z. Metallkunde, 1933, 25, 310— 313).— The textures produced in the inside of metal wires by rolling, drawing, or free stretching are exactly the same, but the outer layers of rolled wires show considerable differences from those of drawn wires, owing to the inequalities in the deformation of various parts produced by the rolling. In drawn wires the sim ilarly oriented crystal axes in the outer layers are inclined at a few degrees to the axis of the wire, owing to the metal of the outer layers having a direction of flow in the die inclined to the wire axis. The X-ray tests were made on copper, aluminium, iron, and j3-brass.— B. B l. *X-Ray Line Broadness of Metals and Alloys, and Its Relation to High- Temperature Stability. J. E. Wilson and L. Thomassen (Trans. Amer. Soc. Metals, 1934, 22, 769-809).— Recovery of X-ray line sharpness in cold-worked metals and alloys is studied as a function of time and temperature of annealing. Line sharpening is caused by diffusion of displaced atoms to normal lattice positions; in the case of nickel at low annealing temperatures the process involves a long induction period. A quantitative time-temperaturo relation ship for line sharpening is determined for nickel, copper, and the a-brasses. A mechanism of recovery is described, based on correlations with hardness, electrical conductivity, and other data. Line sharpening temperatures may bo used to predict immunity from season-cracking in brasses. Creep d a ta on brasses, copper-niekel alloys, and a series of pearlitic manganese and manganese-molybdenum steels are correlated with line broadness data. In a continuous series of solid solutions, the alloy of maximum line-sharpening temperature has the optimum creep characteristics. The poor creep strength of 60: 40 brass above 500° F . (260° C.) is due to inferior temperature stability of th e /3-phase. A secondary m axim um in creep stren g th a t 9 00 ° F . (48 2° C.) in one of the manganese-molvbdenum steels is paralleled by a maximum in a precipitation-hardening effect which is disclosed by X-ray and hardness measurements.— S. G. Various Modes of Attack in Crystallographic Investigations. J. D. H. Donnay, G. Tunell, and T. F. W . Barth (Amer. Mineralogist, 1934, 19, 437- 458).—Descriptive methods are classified as either determinative, morplio- 1935 IV .— Corrosion 17 logical, or structural. The gonio metric methods of Fedorov and of T. W . Barker are reviewed. A system of morphological description, based on tho work of H aiiy and of Bravais, is summarized. The structural classification of crystals is based on the X-ray determination of the space-lattice, and tho present state of rontgcnographic identification is discussed. Cases illustrate, respectively, the agreement and disagreement of tho “ determinative ” lattice with that arrived at by structural methods; it appears, however, that in the case of chemical elements such disagreement is not to be expected.— P. if. C. R .

IV.—CORROSION

♦Weathering oi Aluminium Sheet Used in Aircraft. W illard Mutchler (Nat. Advisory Cttee. Aeronautics, Hep. N o . 490, 1934, 34 p p .; C. Abs., 1934, 28, 722S).— The results of numerous laboratory corrosion and wcather-exposuro tests conducted over a period of more than 5 years, at Washington, D.C., Hampton Roads, V a „ and Coco Solo, Canal Zone, are given in tabulated form, with many illustrations. Tho effects of variable factors in the heat-treatment and the application of various protective coatings arc discussed. Pro perly heat-treated sheet material of tho Duralumin typo is entirely reliable. Protective coatings of pure aluminium are recommended if severe corrosive conditions are to be met.— S. G. ♦Aluminium in Bleaching with Hydrogen Peroxide. R. W . Müller (Dent. Färber-Zeit., 1934, 70, 327-328; J. Textile Inst., 1934, 25, 446a).— 1The d is advantages of the materials generally used for the construction of bleaching apparatus are discussed, and a brief account is given of laboratory tests of the value of aluminium for the construction of apparatus for tho hydrogen peroxide bleaching process. The tests showed that aluminium is only attacked to a very slight extent by dilute ammonia solutions of pu of the order of 8. Commercial hydrogen peroxide solutions which aro slightly acid produce a small loss in weight at first, but further attack is prevented by the formation of a protective oxide film. The addition of dilute ammonia reduces the attack and addition of sodium silicate encourages the formation of the pro tective film. No attack of the metal occurs when the silicate and alkali amounts aro adjusted to give a of 7*5 -9 in solutions containing 1—6(/q b y volume of hydrogen peroxide even when heated to 9 0 °—95 ° C. Determinations of the oxygen loss from hydrogen peroxide solutions stored in different types of vessels showed that the solutions were as stable in aluminium vessels as in porcelain ones, and more stable in these than in glass ones. Large losses of oxygen were observed on storing in iron and copper vessels.— S. G. _ ♦On the Interestin g Sp ecial Case o i the Use o i S ilu m in . G . E c k e r t (Aluminium, 1 034, 17, 3 1 -3 4 ).—Tho effect of boiling linseed oil on aluminium and its alloys has been determined at temperatures up to 3 7 0 ° C. Overheated oil rapidly attacks pure aluminium and most of its alloys, producing severe pitting and eventually perforation; Silum in is, however, quite unattacked even with severe overheating. Boiling stearic, oleic, and palmitic acids arc also entirely without action on Silum in but rapidly dissolve aluminium; boiling phenol has relatively little action, and boiling aniline no action at all on Silumin.—A . R . P. ♦Pipe Corrosion Experiments, Catskill Supply, New York City. Frank h . H a le (./. Amer. Water Works Assoc., 1934, 26, 1315-1347).— F o r 2 years, water was allowed to flow through J-in. diameter pipes of steel, wrought iron, gal vanized steel, Plum rite (6 0 : 4 0 brass), Cuprite (85 : 15 brass), and lead. In a second two-year experiment, pipes of w T O U g h t iron, galvanized wrought iron, copper, and cement-lined steel were used. Seven series of tests w-ere run in each experiment, using raw Catskill water, and water treated with 18 Metallurgical Abstracts sodium silicate to 5 p.p.m. and 10 p.p.tn. of silica, respectively, with o p.p.m. and 10 p.p.m. of soda ash, 4-5 p.p.m. of lime hydrate, and 4 p.p.m. of sodium hydrate Rate of flow was regulated, and, in all pipes except tho lead, lengths maintained at 150°-1C0° F. (65-5°-71° C.) were included in the system. Weekly determinations (twice monthly after several months) were made on tho emerging water for oxygon, free carbon dioxide, value, iron, alkalim ty to phenolphthalein and methyl orange, turbidity, and colour. Every 6 months lengths of piping were removed for examination and total clogging of tho system measured. General averages for each pipe arc tabulated. None of the water treatments rcduccd oxygen consumption in tho cold systems in any pipes, and in the hot systems the silicate treatments alone reduced oxygen consump tion in the wrought iron and steel pipes. The treatments— particularly the s ilic a te treatments— caused worst clogging in the hot wrought-iron pipes, less in black steel, cement-lined steel, galvanized steel, and galvanized w-rought iron (in that order), and practically none in lead, copper, brass, and red brass. Oxygen consumption (and corrosion) is very much greater in the ferrous than in the non-ferrous pipes.— J. C. C. »Rate of Solubility of Gold, Silver, and Copper Alloys in Cyanide Solutions. I. N. Plaksin and S. V. Shibaev {Sovet. Zololoprom, 1934, (3/4), 40-41; 6. Abs.„ 1934, 28, 7230).— [In Russian.] A brief report is given of a study of solubility of the systems gold-copper, silver-copper, and gold-silver in cyanide solution.— S. G. »Printed Matter for Nickel Articles. ---- (Printing Ind. Res. Assoc. Memo. No. 7, 1934, 7 pp.).—Labels printed with a black ink on a glassine paper caused fogging of niekcl-plated articles to which they were attached. Ih e paper contained a trace of sulphur dioxide and appreciable quantities of sulphates. The mechanism of fogging is explained by reference to the work of V ernon (./. Inst. Metals, 1932, 48, 121-136). In order to prevent fogging it is recommended that a thin film of lanolino should be applied to the nickel prior to attaching the label and that 0-1% of ammonium sulphido be added to the solution used to damp the gum at the back of the label, lrinted matter brought into contact with nickel articles should be as free as is feasible from materials likely to give rise to sulphur dioxide, and should contain tho minimum of sulphides, i.e. in the case of paper it should nave as low an acid value as possible.—S. G. *On Some Ancient Copper-Coated Silver Coins of Cyprus. Stanley U. W illim o t t (J. Inst. Metals, 1934, 55, 291-294).— A number of authentic Greek and Roman coins of Cyprus having the appearance of copper or bronze have been found at different times and places. On archasological grounds these coins would have been expected to have been silver, and laboratory investigation proved this to be the case, 92*3% silver being found in one case &pd 94% in another. The cause of this phenomenon was studied and 3 possibilities were considered : (1) galvanic action due to the chance contact of silver coins with a less noble metal, e.g. iron, in the presence of water containing copper sulphate as an electrolyte; (2) the electrolytic (cathodic) deposition of copper on silver as a result of contact with an electrolyte charged with copper salts derived from cupriferous pyrites; (3) chemical alteration of the surface of the silver coins by pyro-oxidation of the contained coppcr as a result of accidental fires. W ith regard to cause (1) it was possible to demonstrate this in the laboratory and to coat silver coins with a tenacious film of copper. —S. G. W. *The Corrosion of Tin and Its Alloys. I.— The Tin-Rich Tin-Antimony- Copper Alloys. T. P. Hoar (J. I7ist. Metals, 1934, 55, 135—146; discussion, 146-147).—The tin-rich tin-antimony-copper alloys were examined with regard to their resistance to corrosion by dilute hydrochloric and citric acids and by various tap-watcrs. The straight 5% antimony alloy containing no 1935 IV .— Corrosion 19

copper is found to bo usually somewhat more resistant to these types of corrosion than alloys containing copper. Soft water produces tarnishing; hard water gives no tarnish, but may give localized attack if there is much chalky deposit.—T. P. H. *A Reflectivity Method for Measuring the Tarnishing of Highly-Polished Metals. L. Kenworthy and J. M. Waldram (J. hist. Metals, 1934, 55, 247— 256; discussion, 256-263).— An optical apparatus and method are described for determining the specular and diffuse components of reflection of light from the surface of polished metals and thereby measuring the degree of tarnishing when such specimens are subjected to prolonged exposure to the a ir. I f Ji3 and Itn arc respectively the specular and diffuse reflectivities, a “ figure of merit ” (M) may be calculated by the expression M = 1 0 V R„(RS + RD)I(RD + 1). For pure tin exposed to the air M decreases almost hyperbolically from 300 to 100 in 16 weeks, for ordinary tin from about 180 to 110, and for Britannia metal from about 320 to 140 in the same tim e; periodic washing in soap and water increases M again but never to the original value, for tin M after each washing returns to 200-250 and for Britannia metal falls slowly to 240. The fall in M produced by washing is attributed mainly to the effect of scratching. In the discussion A . Portemn adversely criticized the “ figure of merit ” as a criterion of the reflectivity of a metal and described a method of measuring the brilliance or polish of a metal by means of the Toussaint photocolorimeter; this method was claimed to overcome difficulties due to colour variations when the metal tarnished, and to be less liable to error sincc observations were made with the photo electric cell. Some results on the colour changes of industrial metals on exposure to air and on the changes in reflectivity thus produced were described and illustrated graphically.—A. R. P. *Some Optical Observations on the Effect of Ozone and Air on Metals. L. Tronstadt and T. Hoverstad (Trans. Faraday Soc., 1934, 30, 1114—1127).- The formation of films on silver, coppcr, zinc, iron, and steel in dry and moist ozone has been followed by Crude’s optical method; in the moist gas no highly protective films are formed on any of these metals, and the attack is partly duo to ordinary atmospheric corrosion, but in dry ozone such films are formed on silver, iron, and steel. The films formed on copper and zinc in dry ozone are not completely protective, no stationary state being attained. The growth of films on zinc in moist and dry ozone and pure air is a linear function of the time of exposure; their formation is discussed from the point of view of recent work on electron diffraction. Film s produced on copper and iron in an atmosphere containing iodine have only a poor resistance to the diffusion of the reagent and hence must be regarded as porous or liable to crack.—A. R. P. On the Corrosion of Welds. Francis Meunier (Bull. tech. Suisse Ramande, 1934, 60, 253-255, 269-272).— A brief, general consideration of the phenomena of corrosion and of the formation cither of a protective film of oxide or of “ passivating ” ions is followed by a discussion of the best methods of securing homogeneity of structure at the weld. Electrolytic potentials and results of impact tests on a series of welds arc tabulated, and impact figures are com pared with the amount of dissolved gas (oxygen and nitrogen) present in the welds. The proportion of dissolved oxygen is found to increase the brittleness of the material. In the case of aluminium, corrosion is comparatively slight if suitably coated electrodes are employed. Stainless steels are then discussed, their remarkable immunity being correlated with the intrusion of chromium into the y-iron space-lattice.—P. M. C. R. 2 0 Metallurgical Abstracts Vol. 2

V.—PROTECTION (Other than Electrodeposition.)

tTechnical Applications of the Eloxal Process. H. Schmitt and L. Lux ( Alum inium , 1934, 17, 191-195).—A review of the chemical and physical properties of the oxide coating on aluminium produced by the electrolytic process and o£ the uses of metal so treated in the chemical, electrical, and other industries.—A. R . P. *On the Prevention of Attack of Water on Light Metals. H. Rohrig and K . Krekeler (Aluminium, 1934, 17, 140-141).— Experiments are described to show that the addition of about 1% of an emulsified oil to water in which light alloys are immersed reduces considerably the rate of corrosion and increases the fatigue lim it by forming a thin film of oil over the metal, thus preventing intercrystalline attack.—A. R . P. Corrosion Protection in Hot-Water Apparatus. Erich Naumann (Z. V .d. I., 1934, 78, 472-476).— Corrosion of hot-watcr apparatus is caused principally by oxygen. The behaviour of steel, copper, and tinned copper as con structional materials is discussed, and various types of apparatus, methods Qf degassing the water, and chemical means for deoxygcnating the water are described.—v. G. ♦Protective Value of Nickel and Chromium Plating on Steel. William Blum, Paul W . C. Strausser, and Abner Brenner (J. lies. Nat. Bar. Stand., 1934, 13, 331-355; Research Paper No. 712).— Exposure tests of plated steel were conducted in co-operation with the American Electroplaters’ Society and the American Society for Testing Materials, in rural, suburban, industrial, and marine locations. The protective value of nickel coatings depends almost entirely on their thickness. The conditions of nickel deposition and of the cleaning have no marked effects on protective value. An intermediate layer of copper decreases the protective value of thin deposits but is not detrimental in thick coatings, especially if they arc chromium plated. The customary thin chromium coatings (0-00002 in. or 0'0005 mm.) increase the resistance to tarnish, but not the protection against corrosion. Relatively thick chromium coatings (0-00005-0-0001 in .; 0-0013-0-0025 mm.) improve the protection against corrosion, especially in industrial atmospheres. The protective value of chromium over nickcl or composite coatings is some what improved by using a bath with a high ratio Cr03/S04 (grm./litre). Deposits produced at 35° C. are slightly superior to those made at somewhat higher temperatures. The use of zinc under nickel makes the protective value less than that of either metal alone. Cadmium has very little effect under nickel. The use of zinc or cadmium under nickel tends to produce white stains and blisters.— S. G. ♦Accelerated Tests of Nickel and Chromium Plating on Steel. Paul W . C. Strausser, Abner Brenner, and W illiam Blum (J. lies. Nat. Bur. Starul., 1934, 13, 519-526; Research, Paper No. 724).— Plated specimens similar to those used in atmospheric exposure tests were subjected to accelerated tests, especially by means of a salt-spray and by intermittent immersion in salt solution.' The time required in these tests for the first appearance of slight rust was not consistent and had no direct relation to the protective value of the coatings. When the extent of rust at the end of a definite period, e.g. 100 hrs., was recorded, the results were approximately parallel to those of atmospheric exposure. The protective value of a metallic coating of this type depends principally on its freedom from porosity. The latter can be determined in a few minutes by the ferroxyl test.-—S. G. 1935 V I.—Electrodeposition 21

»Summary: Protective Value ol Plated Coatings on Steel. W . Blum, T. W . C. Strausser, and A. Brenner (Monthly Rev. Amer. Eledroplaters' Soc., 1934, 21, (2), 8-10).— Seo preceding abstracts.— A. R . P. The Electrolytic Production oi Nickel Matrices. A. Wogrinz (Metallwaren■ Ind. v. Galvano-Tech., 1934, 32, 497-498).— A brief account is given of the production by elcctrotytic methods of letter matrices for printing. A. R . 1 • The Testing oi Tinplate ior Food Preservation. H. Serger and Fritz F le isc h e r {Must. Zeit. Bkchindustrie, 1934, 63, 1165-1107).—Tinplate intended for the sterilization and preservation of food must be soft-, elastic, and non- porous; the depth of tinning should correspond to a deposit per 100 cm.2 of 0-3 grm. on each side, and the lead content should not exceed 1%. The method of testing for brittleness by hammering is described, and the classi- fiqation of its results explained. The thickness of the deposit is measured by a micrometer. A description is given of chemical methods of estimating the thickness, both of deposit and of the bonding layer of tin-iron compound. Porosity tests are also described, and the properties and application of suitable varnishes arc outlined, as are the causes of certain common defects and their influenco on the contents of the vessel.— P. M. C. R. Painting o£ Aluminium and Its Alloys. H. Rohrig and \\. Isicount {Maschinenbau, 1934, 13, 293-294; and (translation) Light Metals Rev., 1934, 1, (2), 19-24).—Various points of view are discussed.— K . S.

VI.—ELECTRODEPOSITION '

Electrodeposition o£ Aluminium and Other Metals. C. L. Mantcll {Metal Cleaning and Finishing, 1934, 6, 397-400; C. Abs., 1934, 28, 7172).— Rccent progress is described.— S. G. ' »Alkaline Plating Baths for Cobalt and Nickel with High Throwing Power. F. C. Mathers, G. F. Webb, and C. W . Schaff {Metal Cleaning awl Finishing, 1934, 6, 412, 418-^19; C. Abs., 1934, 28, 7171).— Sodium-potassium tartrate (Rochelle salt) and ethanolaminc gave complex salts of cobalt and nickel which were soluble in alkaline solutions. The results with cobalt were much better than with nickel, but in no case was the quality of any deposit equal to that from the ordinary, practically neutral, baths. The best cobalt bath contained (in grm./litrc) cobalt sulphate 10, Rochelle salt 100, and sodium carbonate 100.' A current density as low as 0-03 amp./dm.2 is recommended for bright deposits. The cobalt anodes did not corrode; hence the anode efficiency was zero. Cobalt salts were added, therefore, in order to main tain a suitable metal concentration. The bath seemed to bo stable at room temperature but at 70° C. a precipitate formed, causing dark deposits. Operated at room temperature these baths had an average throwing power of +65 at a 3 :1 ratio. The best nickel bath contained (in grm./litrc) nickel chloride 20, ethanolaminc 200, sodium chloride 100, and sodium carbonate 100. The cathode current efficiencies averaged 50% and the throwing power was 40-60 at a 3 : 1 ratio as compared with 0-20 for ordinary nickcl baths. A current density of 0-24 amp./dm.2 was used. The baths could be main tained only by additions of nickel salts because of a gradual precipitation. Further additions of ethanolaminc would not rcdissolvc this precipitate or prevent its formation. Rochelle salt in place of ethanolaminc gave thin, dark deposits.—S. G. _ Copper Electroplating Procedure. W . A. Koehler (Metal Cleaning and Finishing, 1934, 6, 401-404, 419-420).— A review.— S. G. Practical Plating. Deposition of Copper. II.—Acid Copper Solution. E. A. Ollard and J. W . Perring (Met. Ind. (Lond.), 1934, 45, 519-521). the use of phenol-sulphonic acid as an addition agent, and the operation, control, 22 Metallurgical Abstracts Vol. 2 and purification of the solution are described. (Sec Met. Abs., 1934, 1, 596.)-—J . H . W . *The Effect o! Oxidizing Agents on Nickel Deposition. I.—Hydrogen Per oxide and Nickel Nitrate. A. W . Hothersall and R. A. F. Hammond (Trans. Faraday Soc., 1934, 30, 1079-1094).— The effects of additions of pure hydro gen peroxide and nickcl nitrate on the cathode efficiency, hardness, appear ance, cathode potential, and gas discharge at the cathodc have been deter mined in electrolytes of nickel sulphate buffered w ith boric acid or ammonium sulphate. The action of the oxidizing agents is primarily one of hydrogen depolarization, the cathode efficiency being decreased linearly with increasing concentration of the oxidizing agent to an extent which is slightly greater with high than with low pn" The effects of the two oxidizing agents are proportional to their oxidizing power, i.e. 1N03' = 4H202, the N 03' being reduced to ammonia. The hardness, stress, and bright appearance of the deposits are increased by the addition of oxidizing agents, probably owing to the greater amount of colloidal basic substances which are included in the deposit, hence an excess of these agents should be avoided to prevent cracking and exfoliation of tho deposits. As the concentration of oxidizing agent is increased the amount of gas discharged at the cathode decreases to zero over a wide range of concentration, but at higher concentrations gas discharge begins again, except in solutions containing ammonium sulphate and nickel nitrate. The function of hydrogen peroxide in preventing pitting of deposits from impure solutions is discussed.—-A. R . P. The Nickel-Plating of Aluminium and of Light Alloys. Anon. (M ust. Zeit. Blechindustrie, 1934, 63, 1229-1232).— Solutions previously used for the preparation of aluminium and light-alloy surfaces for nickel-plating were so strongly acid as to produce marked surface attack. A hot solution of ferric chloride, acidulated with hydrochloric acid, is proposed as a substitute. Hydrolysis may be prevented by the addition of tartaric acid. Suitable concentrations aro given for the plating of aluminium and of Duralumin, and cleaning and preparation are described.—P. M. C. R. »The Throwing Power oi Zinc Sulphate Solutions. Masami Nakajima (J. Electrochem. Assoc. (Japan), 1934, 2, 176-181; C. Abs., 1934, 28, 7171).— [ In Ja p a n e s e .] Cf- Met. Abs., 1934, 1, 309, 421. The relationship between the current density and the throwing power of zinc sulphate solution was determined for 0-5N solution by the gravimetric and potcntiometric methods, and for 1-5 and 2-5N solutions by potentiometric determination. In every solution the throwing power increases with increase in the current density, and with increasing pa value for solutions of the same concentration. The conductivity of tho bath and the cathodic polarization potential do not greatly affect the throwing pow'er. Thus 1-5 and 2-5 N solutions with a pa of 3-4 gave the highest throwing power. The conductivity of the baths bears a linear relation to its temperature.—S. G. Controlling the Character of Electrodeposits. W . A. Koehler (Metal Cleaning and Finishing, 1934, 6, 345—348; C. Abs., 1934, 28, 7172).— A discussion of the. influence of current density, temperature, ionization, addition agents, and throwing power on the character of electrodeposits.—S. G.

VII.—ELECTROMETALLURGY AND ELECTROCHEMISTRY (Other than Electrodeposition.)

Relation o! Research to the Establishment of Electrometallurgical Industries in the State of Washington. The Development of a Magnesium Industry. Car! F. Floe (State Coll. Washington, Met. Res. Bur. Information Circ. N o . 9, 1934, 5 pp.).— General.—S. G. 1935 IX .—Analysis 23

»Arcs in Air Between Metallic Electrodes. H. Dziewulsld (Acta Phys. Polonica, 1933, 2, 51-58; C. A ls., 1934, 28, 7173).— [In French.] In arcs between similarly shaped electrodes of tungsten, tantalum, molybdenum, copper, nickel, and iron the force opposed to the e.m.f. is independent of, or only slightly dependent on, the arc length, and in the range between 3 and 4 anip. it is independent of tho current intensity. Since this force is equal at both electrodos, irrespective of the combinations studied, it is assumed that it is localized in the layer adhering to the electrodes.— S. G.

VIII.—REFINING

The Electrolytic Refining of Tin. Werner Frölich {Metallbörse, 1934, 24, 826-827, 859).— Tho use of the following electrolytes is briefly described and their advantages and disadvantages are pointed o u t: sulphate, eresolsulphon- ate, fluosilicate, and thiostannate.—A. R . P.

IX.—ANALYSIS

*The Spectrographic Analysis o£ Some Alloys of Aluminium. Ernest H . S. van Someren (./. Inst. Metals, 1934, 55, 265-272; discussion 272-274).-- Describes the techniquo of the analysis of A1 alloys by means of their spark spectra in the ultra-violet, using the method of internal standards. -Tables are given for the estimation of Cu, Zn, Fe, Si, Mn, Mg, N i, Sn, and Cd, and also for the detection of Pb, Sb, Cr, Ti, and Bi. E. H. v. o. Qualitative Spectral Analysis of Pure Metals. R. Breckpot (A atuurw. Tijdschr., 1934,16, 139-143; C. Abs., 1934, 28, 7194).-A previously desenbed method was used with an arc as spcetral source, overheating being avoided. The lim it of accuracy was 0001-0-0001%. In some cases concentration of small amounts of impurities is necessary, e.g. by evaporation with UutAU,), and ignition of the residue. Details are given of the analysis of commercial and electrolytic Cu, commercial Sn, Al, &c.—S. G. Precision Methods for X-Ray Examination of Metals. S. T. Xonobieyskii (Zavadskaya Lab., 1933, (6), 23-28; C. Abs., 1934, 28, 7!94) - [I n lUissian ] A general discussion of the apparatus and a few examples of the application of the method.—S. G. . _ Tr . , n The Application of Radiographic Methods in Chemistry. Otto Hahn (Ber. dent. chem. Oes., 1934, [A ], 67, 150- 163).-Radio-chemistry may be applied to the elucidation of certain processes by the aid of radiographic indications, e, the phenomena of “ inner adsorption during the crystallization of mixtures of constituents of widely differing concentration, of anomalous solid solutions, and of radioactive alloys. A further field of application includes the study of the surface conditions of solids as affecting their retention of radio-emanations.—P. M. C. R. , Microanalysis. H. W . Hermance (Bell Lab. Record, 1934, 13, 81-87).— A general, illustrated account of the special methods used in microanalysis, with particular reference to tho work of the microchemical section of tho

^°*R^searchesCoutlie Limits of the Spectrographic Detectability of Cadmium and Palladium in Silver. Georg Baiersdorf (StUungsber. Akad. Wien, 1934 Ila 143 19-29).—The limiting amounts of Cd and Id which can be detected in Ag vary with the means used in exciting the spectrum Using Sriach ’s method the limits arc : for Cd 0-1 p.p.m., and for Pd a W l Rp .g n. 24 Metallurgical Abstracts Vol. 2

♦Detection of Calcium in the Presence of Strontium and Barium. Earle R. C a le y ( Indust. and Eng. Chem. (Analyt. Edn.), 1934, 6, 445-447).— Tho detection depends on the precipitation of Ca(OH)2 bv the reaction : CaCL + HgO + 4KI + H 20 = Ca(OH)2 + 2KC1 + K 2HgIj; Ba and Sr arc not pre cipitated since their hydroxides are much more soluble.—A. R. P. ♦Compounds of Vanadium with 8-Hydroxyquinoline and Their Use in Analysis. R. Montequi and M. Gallego (Anal. soc. fis. quirn., 1934, 32, 134- 135).— In C1T,C02H solutions V as vanadate gives a violet coloration or pre cipitate with 8-hydroxyquirioline; on shaking with CHC13 the latter develops a characteristic red colour: sensitivity 1-5 ¡xg.V/c.c.— A. R . P. Precipitation of Barium in the Copper-Tin Group of Qualitative Analysis. U illiam T. Hall and Robert B. Woodward (Indust, and Eng. Chem. (Analyt. Edn.), 1934, 6, 478).— In separating As from Ba by H „S the solution should be 0'3Ar with respect to HC1; in the presence of UNO, BaSO. is copre- cipitated with tho As2S3.—A. R. P. 3 1 Separation of Iron from Indium with Cupferron. Frank C. Mathers and Clarence E. Prichard (Proc. Indiana Acad. Sci., 1933, 43, 125-127).—Cup ferron precipitates Fe, Cu, &c., leaving A1 in solution. In is closely related to Al and the same separation may be used. The In is precipitated with XIIjO TI, and tho residue ignited and weighed as ln 20,. Since InCl, is solublo in ether + HC1, the method of Gooch and Havens for the separa tion of Fe and A1 cannot be applied to Fe and In.— R. G. ♦Studies on the Anthranilic Acid Method for the Gravimetric Determination of Metals by Means of the Thermobalance. Saburo Ishimaru (Kinzoku no ICenbtu 1934,11, (10), 500-504).— [In Japanese.] See Met. Abs., 1934, 1, 605.— S . g ! Determination of Antimony in Solder. Clifford L. Barber (Indust, and Edn.), 1934, 6, 443Ht45).—The presence of as little as 0-01% of Sb in solder may be detected by boiling 1 grm. of the filings with 50 c.c. of HC1; if fine bubbles of H , are evolved and the alloy disintegrates bb is present. For its determination 2 grm. of sawings are dissolved in boiling j . exP?lled by boiling with 15 c.c. of II20 and 15 c.c. of HC1, and the Sb determined by titration of the cold, diluted solution with 0-05iV- KMn04.—A. II. P. Determination of Small Quantities of Antimony in White Metals : A Volumetric Method. C. W . Anderson (Indust, ami Enq. Chem. (Analyt. Edn.), 1934, 6, 456-457).—The alloy is dissolved in HC1 and B r and tho solution evaporated with Na2S03 to volatilize As; the Sb is then determined bv titration with K Br0 3.—A. R. P. ♦Reducing Action of Mercurous Chloride. Separation, Detection, and Estimation of Arsenic, Gold, Platinum, Palladium, Selenium, Tellurium, and Gordon G. Pierson (Indust. and. E/u,r. Chem. (Analyt. Edn.), 1934, 6, 437-439).-— .Traces of tho precious metals in a mixed solution may be detected and determined as follows : the solution is boiled with 5% of H 2C204; any precipitate is collected and dissolved in HC1 + Cl2, and the solution, after expulsion of CL, is boiled with Hg2Cl2 which bccomes coloured pink to purple if Au is present. The filtrate from the ILC ,0 , treatment is evaporated to fumes with H 2S04, the solution is diluted with 2-3% HC1 treated with CuS04-5H,0 to give a 5% solution, and shaken with 0-1 grm’ of Hg2Cl2 which precipitates the Pd; the filtrate from this treatment when boiled with Hg,CL for 5 minutes gives a precipitate of Pt. Ignition of the H g,q, precipitates affords the pure metal (Au, Pd, or Pt). Te arid Se can be precipitated in 20% HC1 by warming with HgtCi2, and As can similarly be precipitated from 30% HC1.—A. R. P. ' *The Spectrographic Detection and Estimation of Minute Quantities of Impurnies in Copper. M. Milbourn (J. Inst. Metals, 1934, 55, 275-281; discussion, 281-282).—An accurate and convenient method is described for 1935 IX .—Analysis 25 the detection and estimation of small quantities of Bi, As, Pb, Fe, Ni, Ag, Sb, and Sn in Cu. Details of technique, sensitivity, and line intensity comparisons aro given.—-M. M. *A “ Synthetic Spectrum ” Method of Analysis and Its Application to the Quantitative Estimation oi Small Quantities of Bismuth in Copper. D. M. S m ith {J. Inst. Metals, 1934, 55, 283-289; discussion, 289-290).— A method has been devised for the production of “ synthetic ” spectra as standards of comparison, the spectrum of a standard alloy being exactly superimposed on that of the pure metal -which forms the main constituent. The total time of the 2 exposures is equal to the normal time of exposure of a sample which is being analyzed, and a scries of spectra is obtained in which the impurity lines show a systematic increase in intensity. W hile the method was origin ally applied to the checking of the reliability of standard samples, it can be used for the quantitative determination of impurities in metals, once the standard calibration curve has been obtained. The application to the deter mination of 0-0001-0-004% of B i in Cu is described.— D. M. S. A Volumetric Method for the Determination of Cobalt and Nickel. J. T. Dobbins and J. P. Sanders ( Indust. and- Eng. Cliem. (Anahjt. Edn.), 1934, 6, 459-4G0).—The method depends on the precipitation of the motal -with N H 4CNS and C6H 5N and titration of the excess of the former in an aliquot part of the filtrate with AgN03.—A. R. P. ♦Rapid Potentiometric Method for the Quantitative Determination of Copper in Alloys. Henry B. Hope and Madeline Ross ( Indust. and Eng. Chem. (Anahjt. Edn.), 1934, 6, 316-318).— T h e sam ple co n tain in g 0-1-0-2 grm . of Cu is dissolved in H N 03 and the solution evaporated with 3 e.e. of H 2S04 until copious fumes of the latter are evolved; the cold acid is diluted with 15 e.e. of H 20, neutralized with N HjO H, and poured into 10 c.c. of saturated ILSO j solution. A measured volume of KCNS is added, the solution boiled to expel S02, cooled, and filtered. The excess of KCNS is then titrated in HC1 solution with K I0 3.—A. R. P. Flame Determination of Copper by Carbon Tetrachloride. Peter Gabriel (Indust, and Eng. Chem. (Analyt. Edn.), 1934, 6, 420).—Cu in an alloy may be detected by holding a portion in a Bunsen flame and admitting CC14 into the air inlet of the burner; a blue flame appears even when only traces of Cu are present and none of the other metals interferes.—A. R. P. ♦Quantitative Determination of Lead as Periodate. Hobart H . W illard and J. J. Thompson (Indust, and Eng. Chem. (Anahjt. Edn.), 1934, 6, 425-426).— Pb can be separated from Ni, Cu, Zn, Cd, Al, Ca, and Mg by addition of N alO , to the boiling solution of the nitrates in 0-025iV-HN03. After cooling to 0° C. the precipitate of Pb3H 4(IO 0)2 is collected in a porous filter crucible, washed with ice-cold H aO, dried for 2 hrs. at 110° C., and weighed. Alter natively, the precipitate m ay be dissolved in HC1 and excess of H 3As03, and the excess of the latter titrated with K I0 3 using CHC13 as indicator.—A. R . P. [Estimation of] Lead, Antimony, and Tin in Type Metals and Similar Alloys. Louis Cudroff (Chemist-Anahjst, 1934, 23, (4), 6, 8-9).—The alloy (0-5 grm.) is dissolved by heating with 15 c.c. of H 2S0 4 and 5 grm. of K H S 0 4, and the product is dissolved in 100 c.c. of cold 10% tartaric acid solution. After boiling for 10 minutes and cooling for 2 hrs. the PbS04 is collected and converted into PbCr04 for weighing. The filtrate is evaporated to 50 c.c., treated with 00 c.c. of HC1, toiled 1 hr. to expel As, diluted to 150 c.c., titrated with K B r0 3 for Sb, reduced with a Ni coil, and titrated with K I3 for Sn. — A . R . P . Ctycfohexanol in Colorimetric Determination of Molybdenum. Loren C. Hurd and Fred Reynolds (Indust, and Eng. Chem. (Anahjt. Edn.), 1934, 6, 477-478).—The colorimetric determination of Mo with SnCI2 and KCNS is improved by extracting the coloured complex with cycZohexanoI.— A. R . P. 26 Metallurgical Abstracts Vol. 2

»Determination of Tellurium in Tellurium-Lead and Tellurium-Antimonial- Lead. W. J. Brown (Indust, and Eng. Chem. (Analyt. Edn.), 1934, 6, 4-28- 429).—Te-lead (12’5 grm.) is dissolved in 100 c.o. of H N 03 and tho Pb pre cipitated with 25 c.c. of H .SO j; after dilution to 500 o.c., 400 o.c. are filtered through a dry paper and evaporated until H.SO^ fumes are evolved. The acid is diluted with 50 c.e. of ILO , 5 grm. of tartaric are added, and tho solution is boiled, cooled, and filtered. The filtrate is saturated with H 2S, the precipitate collected, washed, and digested with 1 grm. of NaHC03 and 5 c.o. of 10% Na„S solution, and the filtered solution acidified with H 2S0, and saturated w itli H 2S. The Te precipitate is dissolved in HC1 and KC103 and tho solution evaporated to dryness with H N 03; the residue is dissolved in 100 o.c. of 9: 1 HCl, the solution boiled with 1-2 grm. of tartaric acid, and the Te precipitated with SO,, collected, washed witli ILO , then with C2H 6OH, dried at 100° C. for 1 hr., and weighed. Antimonial-Pb is dis solved in aqua regia containing KCI, tho PbCI2 removed, and tho filtrate evaporated to dryness; tho residue is dissolved in 9 : 1 HCl and the Te separated with S0 2. The precipitate is dissolved in HN0-, and purified as for Te-load. The CuS or B i2S3 residue from the Na,S digestion must bo worked up to recover traces of Te.—A. R. P. Ceric Sulphate for Estimating Tin in Bearing Metals. L. G. Bassett and L. P. Stumpf (Indust, and Eng. Chem. (Analyt. Edn.), 1934, 6,477).— Ce(S04)a with starch-KI indicator may bo used instead of K I3 for titrating Sn in bearing metals —A. R. P. Volumetric Determination of Tungsten. M. Leslie Holt ( Indust, and Eng. Chem. (Analyt. Edn.), 1934, 6, 476).—-Reduction of the \V03 with Zn and HCl, either with or without H sP 0 4, addition of Fe3(S04)2, and titration with KM n04 of tho FcS04 produced, leads to results which are 7-20% too low. No modifications could be found to improve tho results.—A. R . P.

X.—LABORATORY APPARATUS, INSTRUMENTS, &c. (See also “ Testing ” and “ Temperature Measurement and Control.” )

»Autographic Thermal Expansion Apparatus. W ilm cr Souder, Peter Hid- nert, and James Fulton Fox (./. lies. Nat. Bur. Stand., 1934, 13, 497-513; Research Paper No. 722).—Describes an autographic thermal expansion apparatus improved and constructed at the National Bureau of Standards, U.S.A., and indicates tho methods used in its calibration and comparison with the Bureau’s precision expansion apparatus. Tho autographic expan sion apparatus employs tho optical lever method of measuring expansion, and is suitable for many purposes in industrial laboratories. Continuous expansion curves can bo obtained photographically, or they can be observed during the progress of the test. Coeff. of expansion can bo determined from these curves. Transformation regions, if present, can also be located. For ordinary materials such as steel the error of the autographic expansion apparatus is about 6% for the range from 20° to 100° C. and about 3 % for the range from 20° to 500° C. This apparatus does not give as accurate results as the fused quartz tube expansion apparatus also designed at tho Bureau for use in commercial laboratories. The autographic expansion apparatus yields certain information regarding transformation regions not readily obtainable by the other type of apparatus.—S. G. »Study of an Electrically-Maintained Vibrating Reed and Its Application to the Determination of Young’s Modulus. E. G. James and R. M. Davies (Phil. Mag., 1934, [vii], 18, 1053-1086).— An apparatus designed for the study of the transverse vibrations of a reed of dimensions of the order, length 2-3 cm., 1935 X I.—Physical and Mechanical Testing, &c. 27 breadth 0-5 cm., and thickness 0-1 cm., is described. The apparatus is eminently suitable for use in the determination of the values of Young s modulus of short crystals with an accuracy of about 1 part in 1000. The value of Young’s modulus for phosphor-bronze at 25° C. is found to be 1 l ’99g X 1011 dynes/cm.2.— J. S. G. T. *A Simple Apparatus for the Continuous Variation o£ Current, Suitable lor Experiments on Magnetic Hysteresis. R. W . B. Stevens (J.

XI.—PHYSICAL AND MECHANICAL TESTING, INSPECTION, AND RADIOLOGY

♦Elongation Values of Copper and Copper-Rich Alloys. Maurice Cook and Eustace C. Larke (J. Inst. Metals, 1934, 55, 165-184; discussion^ 184-188).— Elongation tests have been made on flat pieces of H.C. copper, 70 : 30 brass, 64 : 36 brass, 80: 20 cupro-nickel, and 95:5 gilding metal. \ »nations in the length of the parallel portion from 1-5 to 8-5 in. has no appreciable effect on the” tx>tal elongation values on 1-in. gauge-lcngths; variations in the thickness between 0-125 and 0-02 in. are sim ilarly without effect on 0-5-in. wide specimens with a 2-in. gauge-length ; with specimens less than 0-02 in. thick the elongation decreases rapidly with decrease in thickness. On specimens 0-06 in. thick the elongation increases with increasing width 28 Metallurgical Abstracts Vol. 2

between 0-25 and 1-5 in., but the effect decreases with increasing gauge- length. A detailed study has been made of the effect of gauge-length on elongation values, and of the distribution of elongation along the gauge- length of all the materials in the soft state, and of 70 : 30 brass after various degrees of cold-rolling. The results for total elongation on gaugc-lengths of more than 0-75 in. agree closely with those calculated by the formula; of U n w in (Proc. Inst. Civil Entj., 1903-1904, 155, (1), 170) and of Krupkowski (Rev. H it., 1931, 28, 529). The value for total elongation is not appreciably affected by the position of fracture provided that this occurs not nearer than 0-5 in. of a gauge-mark. The logarithmic relation between the gauge-length and elongation (Bertella, Giorn. Gen. Civ., 1922, 60, 343) is shown to be invalid. In the discussion, II. O'Neill gave a table showing that the decrease of “ extensibility ” with increasing deformation, i.e. the change in work- hardening capacity for 70 : 30 brass, is very much the same as for H.C. copper. 11. I. Kenyon pointed out that other workers had found that the percentage elongation decreases with an increase in the ratio L/V a ', and constructed a curve showing that the results of C. and L. for 70 : 30 brass given in one part of the paper agreed with this, whereas others did not ; he suggested that minute variations in thickness might account for this.— A. R . P. “ Comparison of Single-Step Long-Time Creep Results with Hatfield’s Time- Yield Stress. A. E. White and C. L, Clark (Trans. Amer. Soc. Metals, 1934, 22, 481—4-94; discussion, 494-504).—Bata are presented to show whether or not one of the short-time methods advocated by certain metallurgists for determining creep characteristics of metals at elevated temperatures yields results comparable to those obtained from a carefully conducted long-time creep test. Results are given for 3 steels at 850° P. (454° C.) and 11 at 1000° F . (538° C.). It is concluded that while the time-yield method does not give results which are in exact agreement with those from the long-time test, it docs offer possibilities as a qualitative test for classifying, for example, a series of steels of a given type, at any temperature.—S. G. *New Methods of Testing by Impact. John Hardie Lavery and Richard V y n n e So u th w e ll ( Inst. Civil Enq. Selected Eng. Papers, 1934, (142), 1-35).— The “ Oxford ” impact machine is described, and an account given of its develop ment. The “ hammer ” is suspended by wires to avoid errors through trans mission of energy by the anvil-spceimen-hammer system to earth. Gentle release is effected by an electro-magnet, and the extreme position determined by a thread passing between felt pads and attached to the hammer. A cylindrical specimen, notched with a concentric groove and supported at its ends, is used. The blow is struck by two projections acting on each side of the notch to'give “ four-point loading.” W ith this machine, plastic deformation is localized, and the energy absorbed is found to be almost exactly proportional to the area of fracture. The influences of root radius and depth of notch, eccentric setting of voke, and span of support were studied by tests on “ Super VNCA ” steel.—,!. C. C. Plioto-Elasticity Apparatus Embodies Novel Features. Elm er K. Timby (Eng. News-Record, 1934, 113, 047).— A universal straining machine developed at Princeton University for straining clear Bakelite models for photo-elastic work can be used for torsion, compression, bending, or shear, w ith a maximum total load of 1000 lb. Centralized control of all adjustments to the loading and optical systems is a feature of the equipment.—J. C. C. •{■Materials Testing M achines. C. H . G ibbons ( Baldwin Locomotives, 1934, 13, (1 ), 24-31; and Instruments, 1934, 7, 223-226).— An illustrated historical review of the development of tensile, compression, and transverse testing machines from about 1638 (Galileo) to 1817 (Peter Barlow).— J. C. C. 1935 X III.—Foundry Practice 29

XII.—TEMPERATURE MEASUREMENT AND CONTROL

The Pneumatic Pyrometer. A Variation of the Air-Gauging Process Applied to Pyroinetry. Anon. (Met. Ind. (Lond.), 1934, 45, .318).— T h e m ethod of measuring furnace temperatures by pneumatic gauging is described. The principle of the method is that if a constant air pressure is connected to ft pipe with an orifice at each end, a pressure w ill be set lip in the pipe which w ill be a measure of the difference in the temperature of the air at the two orifices. In the apparatus described, one of the orifices is situated in the furnace whose temperature is required, and the pressure difference scale is calibrated to give the temperature direct.—J. H. W . Measurements o£ High Temperature. Karl Stein (Natnrforscher, 1934, (11), 403-409; C. /16s., 1934, 28, 0595).—A description is given of various pyrometers in use and of the Ardometer for measuring total radiation.— S . G . New Pyrometer for Very Hot Gases. E. 0. Mattocks (Metal Progress, 1934, 26, (5), 30-33).— The spectral-line reversal method of gas pyroinetry consists in the injection into the hot gas of a small amount of sodium chloride. A tungsten band lamp is so adjusted that the sodium lines in the hot-gas sp e c tru m disappear against the spectrum from the calibrated tungsten filament, the temperature of which is registered by an optical pyrometer.— P. M. C. R .

X III.—FOUNDRY PRACTICE AND APPLIANCES

»Effect of Melting Conditions on the Running Quality oi Aluminium Cast in Sand Moulds. A. I. Krynitsky and C. M. Saeger, Jr. (J. lies. Nat. Bur. Stand., 1934,13, 579-588; Research Paper No. 727).—The effect of maximum heating temperature on the running quality of liquid metal has been studied for 2 grades of aluminium and an aluminium-8% copper alloy. The running quality was measured in terms of a length of a spiral casting obtained by pouring the metal under carefully controlled conditions. The effect of various treatments of the metal was also investigated. It was found that the relation between length of tho spiral and the pouring temperature is linear. The running quality of the metals which had been heated to a maximum temperature of 850° C. was in all cases less than when the metal had been heated to 750° C. Pure aluminium (99-8%) was found to have markedly higher running qualities than commercial (99-2%) aluminium. Tho running quality of the aluminium-8% copper alloy was not much different from that of commercial aluminium.—S. G. Shrinkage Cracks in a Light Metal Casting Caused by the Casting-In of a Bearing Bush. M. Widemann ( Giesserei, 1932, 19, 332-333).— X-Ray photo graphs arc given of a bearing composed of a light metal case into which a keyed- in bush of another light metal alloy has been cast (no details arc given of the composition of either alloy); they show that during cooling of the inner alloy shrinkage cracks arc produced in the neighbouring parts of the case, and there fore that, even with careful work, the foundry cannot guarantee the life of such castings under high-frequcncy vibrational stresses.— A. R . P. Methods of Manufacture and Their Influence on Design. Aluminium Alloys—Wrought and Cast Parts. W . C. Devcrcux (Met. hid. (Lond.), 1934, 45, 489—1-9.'!, 513-516).— A paper (slightly condensed) read before the Scottish Local Section of the Institute of Metals. A discussion of the effect of feeding, heat-treating characteristics, precautions to be taken in high-temperature heat-treatment, casting and grain growth, test results, the influence of various treatments and the effects of cold-working on those results, with respect to wrought and cast aluminium alloys.—J. H. W . 30 Metallurgical Abstracts V o l. 2

1'The Practice of Aluminium Melting. A. von Zeorleder (Aluminium, 1934, 17, 196-201).—A critical review of modem practice in melting aluminium and its alloys, and in the prevention of contamination by impurities introduced by the atmosphere of the furnace, the fuel, and the container.— A. R. P. Production of Ternary Aluminium-Copper Hardeners. Edmund R . Thews (Metallurgist (Suppt. to Engineer), 1934,10, 162-163).—The relative merits of different methods of preparing aluminium-copper-manganese and aluminium- copper-iron hardener alloys are discussed. The method recommended is to add the manganese or iron to molten copper previously deoxidized with phos phorus. The melt is then allowed to cool and is poured into molten aluminium which has been superheated to about 800° C. and covered with a cryolite flux. — R . G. The Moulding o! Aluminium-Bronze and Method of Control of the Quality. C . M eigh (Bull. Assoc. Tech. Fonderie, 1934, 8, 180-185; discussion, 185- 186).—Casting in sand, metal, or mixed moulds is compared and contrasted. Methods of control of the quality of castings are discussed in detail. Control by micrography is strongly advocated and examples given of the estimation of the mechanical properties of a piece from its microstructure. The dis advantages of the separately and integrally cast test-pieces (for mechanical test) are discussed and the use of coupons suggested.— R . B. D. Casting Under Pressure. M. F. Nielsen (Bull. Assoc. Tech. Fonderie, 1934, 8, 106-115).— The various types of machines, their design and construction are discussed together with the different types of nozzle. Gate design, run ning, and air removal methods are reviewed.— R. B. D. Brass Die-Castings. Anon. (Metal Progress, 1934, 26, (5), 26-29).— Somo developments in die-casting practice are summarized. The alloys most generally employed arc a modified Muntz metal (copper 60, tin 1, lead 1, aluminium 0-1%, remainder zinc) and “ W ebcrt” alloy (copper 81-5, silicon 4-25, manganese 0-15%, remainder zinc), and the physical properties of both in the die-cast condition are described. The composition of die steels is discussed, a readily carbonizing type containing up to 7-5% of both tungsten and chromium being recommended.—P. M. C. R.

XV.—FURNACES AND FUELS

Fuel-Fired Furnace Temperature and Atmosphere Control. G. H. Barker (Metallurgia, 1934, 11, 29-32).— The performance of heat-treatment furnaces is determined by the method of control adopted as well as by their suitability, design, and construction. The method of control affects the furnace atmo sphere, combustion efficiency, and consequently the fuel combustion, as w ell as determining a uniform temperature or pre-determined time-temperature cycle. Methods of control, including two-position, three-position, balancing, and float ing control, are fully discussed in the light of modern developments and as applied to different designs of furnaces, and the future trend of furnace control is also considered.—J. W . D. Theory of Regulation of Industrial Electric Furnaces. Max Lang (Elektro wärme, 1934, 4, 201-208; C. Abs., 1934, 28, 7167).— Cf. Met. Abs., 1934, 1, 314. Principles of calculation of temperature control as determined by the prevailing practical operative conditions are developed, and the application of the derived formulae is illustrated by a few examples.—S. G. Continuous Electric Furnace for Strip and Wire. R. Granzer and L. M oennich (Elektrowärme, 1934, 4, 181-184; C. Abs., 1934, 28, 6639).— Principles of calculation of annealing furnace capacity and curves of pro duction for a given width of strip or diameter of wire, and for a given length of furnace are developed.—S. G. 1935 XVIII.—Working 31

»Temperature Distribution oi a Gas Flowing Through a Furnace. K . F. H erzfeld (Physics, 1933, 4, 302-365).— The temperature distribution in gases flowing through a cylindrical fumacc at low pressure and high speed is inves tigated mathematically. The flow is laminar, and two cases are distinguished. Temperature equilibrium is attained throughout when the pressure gradient is loss than 0-003 mm. of mercury per cm. For pressure gradients about 10 times as great only a thin layer of gas close to the furnace wall attains temperature equilibrium.—J. S. G. T.

XVI.—REFRACTORIES AND FURNACE MATERIALS

Suitable Packing oi Refractories. C. Koeppcl (Tonind. Zeit., 1934, 58, 1110-1111).— Refractory bricks should be stacked in railway wagons in such a way that their broadest side is transverse to the direction of travel; bricks of unusual shape having no parallel sides and which can easily move diagonally in the truck should be stacked in trucks the sides of which can be well bolstered. — B . B l.

XVII.—HEAT-TREATMENT

»Experiments in Wire-Drawing. Part IV .— Annealing oi H.-C. Copper Wires of Varying Hardness-Elongation Values. W . E. Alkins and W . Cartwright (J. Inst. Metals, 1934, 55,189-197; discussion, 197-199).—The results of deter minations of the elongation values of a scries of wires of very pure high- conductivity copper, drawn with widely varying amounts of reduction and annealed at varying temperatures for varying periods of time, arc recorded in full and briefly discussed.—W . E . A.

XVIII—WORKING

»Tests on Deep-Drawing. L. Herrmann and G. Sachs {Metallwirtschaft, 1934, 13, 687-692, 705-710).—The Wazau deep-drawing tester is described. The influence of the rounding of the tool and of the grip of the blank-bolder on the deep-drawing value has been determined on 63 : 37 brass, and the factors which cause folds in the cup have been investigated.—v. G. »Researches on the Detection of Copper Flecks on Rolled Aluminium W ire. H. Rohrig (Alum inium , 1934, 17, 149).—-Specks of copper which have been pressed into the surface of aluminium alloys during working can be detected by holding the metal first in the vapour of strong nitric acid, then in ammonia, a blue colour being produced around the copper specks. A quantitative deter mination may be made by immersing the metal for 5 minutes in nitric acid (d 1-41), w h ereb y a ll th e copper, b u t p ra c tic a lly no alu m in iu m , is dissolved, and then determining the copper in solution electrolytically or colorimetrically. — A . R . P . -¡Influence of Friction on the Flow of Metals in Rolling. Erich Siebcl (Stahl u. Eisen, 1934, 54, 1049-1056).—A discussion of the effect of the coeif. of friction on the deformation occurring during rolling of metals, including copper, lead, and aluminium. Cf. Siebel and Osenberg (Mel. Abs., 1934, 1, 459), Pomp and Lueg (Mitt. K.-JV. Inst. Eisenforsch., 1933, 15, 81), L u e g and Osenberg (Mitt. K.-IV. Inst. Eisenforsch., 1933, 15, 99).—W . H.-R. "(■Machinery in Rolling-Mill Construction and Its Maintenance. Erich H o w a h r (Stahl u. Eisen, 1934, 54, 1101-1108).— A review of recent develop ments in rolling-mill plant in Germany with special reference to devices for 32 Metallurgical Abstracts V o l. 2

cooling and lubrication, and for the movement of the rolls relatively to one another. Detailed drawings arc given, and H. discusses the effect of friction and lubrication on the maximum safe pitch of screws which must not move under pressure.— W . H.-R. The Effect of Some Mill Variables on the Gauge of Sheet Brass. C. K . S k in n e r (Met. Ind. (Land.), 1934, 45, 343-344).— The influences of the many variables in m ill operation on the gauge of sheet brass as rolled on the ordinary 2-high null aro set out.—J. H. W . 1 A t a ^ W ^ a n d Oils [Therefor]. Karl Krekeler (Aluminium, , ’ 'r , '-—Suitable lubricants are described for rolling, extruding, decn- drawing, wire-drawing, and turning aluminium and its alloys.— A R 1’ Hard Metals in the Machining of Metals. U . Pirani (Anz. Masch., 1034, 56, 4 5). Examples of the test cutting speeds and size of chip are given for *m^13 metal tools cutting some of the common metals —B Bl W w p m °! ^ ,rninS ^ooIs a* Iufluenoed ^ Shape. 0. W. Boston and 5G7-576) —S°rG ‘1931> 22, 547-567; discussion,

XIX.—CLEANING AND FINISHING

Treatment of Aluminium Ware in the Household. Research Dept, of the Vercmigte Aluminium-Werke A.-G. (Aluminium, 1934, 17, 34-36).__ Methods of cleaning domestic aluminium ware are given; various German proprietary clcansers are recommended.— A. R . P. tCol(™ fnd Surface Protection of Metals by Chemical Processes. H. K ra u se (Maschmenbau 1934, 13, 487-190).— Tables, references, and practical

S u u m , Z ¿ v e m - K S ° f COl° U1'in g eSPCCiaUy C° PPC1'’ Zin° ’ and 101i°i0^ gnoi A'u,min.ill.m- K “rt Vollrath and Georg Lahr (Aluminium, '. f t : J7, 91-92).— Aluminium and its alloys (free from copper) can l)e coloured with various dyes by immersion for J hr. at S0°-90° C „ in a 2’5 % solution of potassium sulphide containing an appropriate dye; thus morin gives a yellow, alizarin a red, a mixture of morin and alizarin a bronze, and vanadyl sulphate a black to brown colour. Mixtures of these reagents with or without bichrom ate produce intermediate tints of brown to orange. A golden-yellow colour is Uined by immersing the metal in 2% permanganate solution; the colour may be darkened by addition of copper sulphate. Recipes are given for 11 colouring solutions.—A. R. P. 6 Researches on the Production of Brown Colours on Copper and Copper Alloys. H. Krause ( Metattwaren-Ind, u. Galvano-Tech., 1934, 32, 453-454, 471-472).__ ° i rccent work on the use of permanganate and chlorate baths with the alloys1*— A j t p°US m SaltS f° r ProducinS hrown tones on copper and its

XX.—JOINING

Atemlntam Structural Work. A. J. Field (Light Metals Rev., I, 1U L-Iu.i).— Brief review of current American practice.— R, B D Metal Rivets- ---- ifeerledcr (Aluminium, I' " *140).-—Io obtain the best mechanical properties in rivets of JJuialum m and similar alloys careful control of the annealing temperature is essential, the best results being obtained by quenching from 500° to 510° C. several small electric furnaces for controlling this operation accurately are illustrated and briefly described. A salt-bath is recommended for obtaining even heating.—A. R. P. ° 1935 XX.—Joining 33

tPractice o£ Light-Metal Riveting. W alther Zarges ( Alum inium , 1934, 17, 127-135,177-188).— Various types of light-metal rivets are illustrated and their properties and uses are described. Suggested standard specifications (dimen sions and mechanical properties) arc tabulated and directions are given for using the rivets in joining articles of aluminium and its alloys; 30 references are appended.—A. R. P. *H °t Riveting o£ Hardenable Aluminium Alloys. R. Irmann (Aluminium, 1034,17,189-190)— Experiments have been made to show that hot riveting of Duralumin and sim ilar alloys is not to bo recommended since, even with air- or water-cooling, the area round the rivets is softened by the heat and widening of the hole is liable to occur under conditions inducing stress. Where the rivets arc so large that cold riveting is unsatisfactory it is recommended that a larger number of small rivets should be substituted rather than use hot riveting.

*How Should Aluminium Vessels for the Transport and Storage of Nitric Acid be Welded? G. Eckert (Aluminium, 1934, 17, 84-87).—A scries of corro sion tests on autogenously-welded joints of 99-5% aluminium has been made to find the best procedure for preventing attack by nitric acid. Plain welds and hammered joints are very rapidly corroded but, after a thorough annealing by the flame of the welding torch, avoiding local overheating, a high degree of immunity from corrosion is obtained.—A. R. P. The Arc-Welding of Copper. Wilmer E. Stine (Machine modernc, 1934, 28, 561-562).—The welding of copper, employing carbon electrodes, may bo effected by the addition either of pure copper or of phosphor-bronze, and both methods are described, with modifications according to dimensions. For the welding of very thin sheet, a phosphor-bronze electrode is recommended, though temperature control presents some difficulty. Certain examples of arc-welded copper are illustrated and described.— P. M. C. R. Fusion-Welding of Copper. Anon. (Soudure el Oxy-Coupage, 1933, 10, 206).— Recommendations in technique.—H. W . G. H. The Key to the Problem of Fusion-Welding Copper : Cuprous Oxide and Its Eutectic. Anon. (Soudeur-Coupeur, 1933, 12, (11), 1-4).—The formation of cuprous oxide eutectic during the welding of tough-pitch copper is said to have been announced for the first time by Lo Grix in 1922. An extract from an article published by him at that time is reproduced, in whioh the phenomenon is explained at length.— H. W . G. H. The Progress ol Electric-Resistance Welding. J.-E. Languepin (Metal lurgy et la Construction mécanique, 1934, Special No. (Nov. 17), 44-45).__ Modern developments in the design of machines are reviewed; the tendency is shown to be tow'ards the use of high power and short time of application. —H. W . G. H. Metallic Arc Welding. H. V. Stead (Marine Eng., 1934, 57, 320-323).— Some of the factors involved in ensuring sound welding are critically discussed, and although referring essentially to structural steel welding are of wider interest. _ Correct welding technique and tho provision of suitable electrodes are considered necessary for sound metallic arc welding. Heat-treatmcnt, where necessary, should bo considered in the light of the process used, the electrode, the form of join, and the shape and size of tho structure. The chief causes of unsound welds are summarized and the variety of faults arising froni any one cause or combination of these causes, are dealt with. Reference is also made to the shrinkage stresses that may exist in a weld and the adjacent metal and to the importance of eliminating such stresses. The inspection of welds and testing procedure to ensure sound and reliable welds is also dealt with.—J. W . I). 34 Metallurgical A bstracts Vol. 2

XXI.—INDUSTRIAL USES AND APPLICATIONS

*The Hygienic Significance of Aluminium Cooking Vessels. Joseph W iihror (Arch. Hyg. Bakt., 1934, 112, 198-216; C. Abs., 1934, 28, 7371).— Analysis of foods that have been cooked in aluminium vessels shows that tho amount of aluminium transferred is far below concentrations which might have any harmful effect.—S. G. Use of Aluminium in Bleaching Apparatus. H. Tatu (Rev. gen. mat. col., 19 3 3 , 37, 4 2 7 - 4 3 6 ; J. Textile Inst., 1934, 25, 132a).—Discusses tho faults of tho usual materials used in tho construction of apparatus for the bleaching of textiles by tho peroxide process and indicates the requirements of a suit able material for this purpose. These requirements are shown to be fulfilled by aluminium. Some practical applications of this metal in the construction of apparatus used in bleach works are described.— S. G. A New Light Aluminium Alloy [Hiduminium R.R. 53B]. ---- (Textile Expl., 1934, 11, (83), 10; J. Textile Inst., 1934, 25, 202a).— Hiduminium R.R. 53B contains copper 2-5, nickel 1-5, magnesium 0-8, iro n 1-2, and silico n 1-2%. Its physical properties in various conditions are given. It is stated that this alloy is proving of value for fast-moving parts, and is being applied in the textilo industry to such components as levers, treadles, brackets, &c.— S. G. Progress in the Production and Use o£ Light Alloys. 0. C. Hodgson (Metal- lurgia, 1934, 11, 37-38).—The recently developed alloys of aluminium and magnesium arc discussed and the properties which render them suitable for various applications are considered. The alloys dealt with are both casting and wrought alloys, and include tho alloys of aluminium with silicon, magnesium, or cerium, the Y alloy, and the “ R .R .” scries of alloys. Special attention is called to the advantages accruing from the hcat-treatment of cast alloys, and to tho importance of determining the fatigue-strength and tho relative corrodi- bility of such alloys. The increasing application of those light alloys in transport, engineering, and marine work is also referred to.— J. W . D. Aluminium for Roof Coverings. R. Haefner (Metallwirtschaft, 1934, 13, 786-789).—Laboratory and practical tests have shown that aluminium forms a satisfactory roof covering provided it is laid properly; a bitumen under-layer is recommended.— v. G. Aluminium in Architecture and Building Construction. W . Zarges (Light Meials Rev., 1934, 1, 25-28).—Translated from Metallbörse, 1934, 24, 683, 697-698, 729-730. See Met. Abs., 1934, 1, 528.— R . B . D . Aluminium Roof on the Radio Corporation of America Building. A. J. F. (Light Meials Rev., 1934, 1, 29-30).—Brief description of the method of crection of the aluminium roofs of a portion of tho building.— R . B. D. Developments in the Uses of Light Metals in the Railway F ie ld . ---- (Light Metals Rev., 1934, 1, 150-191).—Detailed survey of developments during the last few years in the various branches of the railway field, with abstracts of all the more important papers and articles published and an ex ten sive bibliography.— R . B . D . A New Aluminium Train. Anon. (Light Metals Rev., 1934, 1, 1-3).— A short description of a 5-car train for the Brooklyn and Manhattan overhead railway in which heat-treated aluminium-alloy sections and castings were extensively used.—R . B. D. Aluminium in Electrotechnics. H ; Schmitt and L. Lux (Aluminium, 1934, 17, 5-15).—A detailed account is given of the uses of aluminium in the electrical industry with especial reference to the manufacture of free transmis sion cables for long distance distribution of electricity. Details are given of the construction of these cables and their properties compared w ith those of copper. — A . R . P . 1935 X X I.—Industrial Uses and Applications 35

* Aldrey lor Electrical Transmission Lines. ---- Irmann and W. Muller ( Alum inium , 1934, 17, 142-147).—The possibility of substituting Aldrey for copper for overhead conductors for tramways, &c., has been investigated, e sp ecially w ith a v ie w to finding a su itab le co n tact m a te ria l w h ich does n o t w ear the wire unduly or produce an excessive amount of sparking. Satisfactory results have been obtained with carbon and with a copper-aluminium alloy provided that adequate lubrication is used. W ithout lubrication only carbon gives satisfactory results.— A. R . P. A New Clip for Free Transmission Lines [of Aluminium]. M. Preiswerk ( Alum inium , 1934, 17, 147-148).— A clamp, made of Anticorodal, for holding aluminium or Aldrey wires to the supports is illustrated and briefly described. - A . R . P . Heat-Losses in Sugar Mills ; Insulation and Aluminium Applications. K. Sandera (LUty Cukrovar., 1934, 52, 361-368; 0. Abs., 1934, 28, 7054).— Heat losses in sugar mills were computed for insulated and non-insuiated surfaces; those from vacuum pans were 66 kg.-cal./m.2/hr., from evaporators 83-3, from pip in g c a rry in g condensed steam 102, from pip in g ca rryin g superheated steam 177. In a heat balance, the losses from the evaporator were 0-25% of the total; losses from all pipes were 2 % of the total. After insulating about 95% of the piping with insulating material having a heat- flow of 0-08 kg.-cal./m.2/hr., the total heat-losses from pipes were decreased from 2 to 1%. After applying an aluminium film to an insulated surface, the heat-losses from that surface were decreased 10%; when the aluminium film was applied to an uninsulated surface, the dccrea.se in heat-losses was 25%. Tho largest heat saving was effected by spraying aluminium on unin sulated surfaces. During a 6-hr. period the heat-loss from the evaporator surface ranged from 73-8 to 84-6 kg.-cal./m.2/hr.; after spraying the surface with aluminium, the heat-flow did not exceed 71-8 kg.-cal./m.2/hr. The atmosphere in the mill, however, makes it impossible to predict how long the new aluminium film will remain so effective.—S. G. Aluminium Foil and Its Uses. E. Herrmann ( Alum inium , 1934, 17, 15-19).— Besides its use -is a packing material for foodstuffs, tobacco, tea, &c., aluminium finds extensive use as a heat insulator, for decorating the interiors of railway coaches, &c., for lining ship’s refrigerators, for the manufacture of clectric condensers, and for many radio purposes. Examples of these uses are illustrated.— A. R . P. Wrapping of Apples in Aluminium Foil. E. Herrmann (Aluminium, 1934, 17, 190).—Aluminium foil has proved to be a satisfactory wrapping material for preserving apples, provided that the fruit is perfectly dry before wrapping. For a short period after gathering, apples exude a small amount of moisture by sweating and if wrapped before this has dried off the fruit rapidly rots; once this has dried, however, an aluminium wrapper will preserve the fruit until well after Christmas.—A. R. P. Cadmium Poisoning in the Industry. H. Fiihner and W . Blume (Arch. Gewcrbepath. Gevoerbehyg., 1934, 5, 177-184; C. Abs., 1934, 28, 6881).— Cadmium poisonings of recent times are discussed. They are due to the inhalation of cadmium, cadmium dust, and cadmium oxide fumes. Various symptoms are described. Whisky or milk are good antidotes.—S. G. Poisoning by Wine Containing Cadmium [Derived from Plated Filters]. P. Fortner (Pharm. Zentr., 1932, 73, 769-774; Brit. Chem. Abs., 1933, [A], 91).—The wine contained cadmium, derived from the plating of filters through which it had passed.—S. G. Poisoning by Cadmium in Coffee [from Galvanized Kettle]. C. Griebel and F. Weiss (Pharm. Zentr., 1931, 72, 689-690; Brit. Chem. Abs., 1932, [B], 46).—Eleven factory workers were seized with nausea and vomiting shortly after drinking coffee prepared with water boiled in a 50-litre kettle which had been treated with concentrated hydrochloric acid to remove scale. 36 Metallurgical Abstracts V o l. 2

Only the lower part of tho inner surface had been cleaned. The coating of the galvanized iron contained zinc 91-5 and cadmium 8-5%. The coffee w as fa in tly acid and contained cadm ium 0-08, zinc 0-006, ch lo rin e 0-129 and mineral matter 0-25%. Although tho kettle had been rinsed with alkali followed by hot water, sufficient zinc, cadmium, and chlorine had been ab sorbed by the scale on the upper part of the inside to contaminate the coffee to the extent indicated.—S. G. fEstim ation of the Toxicological and Hygienic Effects of Prolonged Absorp tion of Small Amounts of Copper in the Human Organism. ---- Spit.ta ( Reicligesundheitblatt, 1932, 7, S6 2 ; U.S. Public Health Eng. Aba., 1934, 14. W- 42) May, 26; C. Abs., 1934, 28, 7388).—A critical study of the literature of the effect of copper on health. Largo amounts of copper are poisonous and destroy tho mucous membrane of the alim entary canal; smaller amounts, taken over a long period accumulate in the liver and may be related to certain liver diseases. Very small amounts of copper are necessary for normal biological processes. Approximately 4-5 mg. of copper is ingested in the food of a normal adult daily. The minimum amount of copper which affects taste in drinking water is 2 mg./litre. Ordinarily harmful amounts of copper arc not dissolved by wator from copper pipes, but pipes constructed from other materials are recommended.—S. G. Properties and Uses of Lead-Base Bearing Metals. E. T. Richards (Met. Ind. (Lond.), 1934, 45, 533-535).— The disadvantages of lead-baso bearing alloys are largely duo to their being handled according to the technique of tm-base alloys. I lie brittleness of the harder lead-antimony alloys (more than 13 /q antimony) can be overcome by adding a little tin. The alloys are greatly affected by the pouring temperature, which should be 400° G , and the preheating temperature of the moulds, which should bo 200° C. Details of pouring and the precautions to be observed during cooling are set out and the compositions of several of these alloys are tabulated.— J. IL W . *The Clean-Up of Various Gases by Magnesium, Calcium, and Barium. A. L. R eim an n (Phil. Ala//., 1934, [vii], 18,1117—1132).— Characteristics of the getter- ing (clean-up) of residual gases, oxygen, nitrogen, hydrogen, carbon monoxide, and carbon dioxide, in evacuated vessels by magnesium, calcium, and barium are investigated experimentally and the results arc discussed.—J . S. G. T. Some Notes and Observations on Petrol and Diesel Engines [The Cracking Of White Metal in Diesel Engine Big-Ends]. H. R. Ricardo (Proc. Inst. Automobile Eng., 1932-1933, 27, 434-151; discussion, 452-i77 ; J. Hoy. Aero nautical Soc., 1933, 37, 509-522; discussion, 523-546; Diesel Engine Users A-sfoc. 1933; and (abstract) Mech. World, 1933, 93, 283-285).—From a review of the experience of numerous British and Continental manufacturers of Diesel engines for heavy road cars it is concluded that the design of the engine is of relatively little importance in determining the life of the white- metal big-end bearings. Engine speed and lubrication also seem to bo without effect. It does, however, appear to be preferable to emplov very thin whitc-metal linings of the order of 0-01—0-02 in. thick carried in separate steel shells. Where the design and cost permit of the use of a hardened crankshaft, lead-bronze, gun-metal, or Duralumin bearings are much more satisfactory from the cracking point of view but may cause difficulties by seizing when the oil supply fails. Several theories to explain the crackin'' of whitc-metal bearings are put forward and briefly discussed. In the dis° cussion, 1). 11. Pije stated that cracking could be reduced by stiffening the shell to make it less flexible, and 0. S. Wilkinson said that by providin" efficient oil-cooling so that tho temperature of the bearing did not rise above a critical value cracking is almost entirely prevented.— A. R . P. Metallurgy and Uses of Zinc.—I. Stanley Robson (School Sci. Rev., 1934, 16, 21-31).— Adapted from an address to the London Local Ssction of the Institute of Metals.—S. G. 1935 X X III.—Bibliography 37

A Brief Discussion of Practical Application oE Metals in Naval Machinery Design. C. S. Gillette (./. Amer. Soc. Naval Eng., 1034, 46, 480-495).__ A survey of the properties of metals as they affect the engineer deals with their selection for naval design with special reference to their resistance to corrosion and erosion, their maximum resistance to failure by shock or impact, their minimum cost, and their maximum availability within the industrial organizations of their own country. The non-ferrous metals referred to are the white metals and the leaded bronzes for bearings, and the various aluminium alloys for lightly stressed or unstressed parts, where light ness and a fair resistance to corrosion are required.—J. W . I). Types of Tube Joints. G. Elders (Z.V.d.I., 1934, 78, 460-465).— Various types of joint are described and their application to steel, copper, aluminium, and lead tubes is discussed.— v. G. German Technique and Raw Material Economy. Anon. (Z.V.d.I., 1934, 78, 1285-1290).— The substitution of aluminium and zinc alloys and artificial materials for copper and its alloys is discussed.— K . S.

XXII.—MISCELLANEOUS

The Work of Walter Rosenhain. J. L. Haughton (J. Inst. 3Ietals, 1934, .» 17-32).~-Autuinn Lecture. A very full bibliography/ is given of R.’s scientific publications.— S. G. ^search on Metals and Alloys. F. C. Frary ( Indust. and Eng. Ohem., 11 ■ —Read before the American Chemical Society. A very able review of the two fundamental objectives in research on metals and alloys - the academic and the practical—and of the fields which have to be covered by the research-workers in order to make progress in the development of new and useful processes and products in the metal industry. The problems discussed are grouped under the heads : effect of impurities, workability, melting problems, customers’ problems, casting processes, corrosion problems. — F . J . The Story of Early Metallurgy. X I.—Copper and Bronze : The Question oi Priority and the Mineral Sources. R. T. Rolfc (Met. Ind. (Load.), 1934, 45, 485-488).—The claims of Sumeria and Egypt to be the first smelters of copper and users of bronze, and the sources whence these countries derived their supplies of copper, tin, and nickel are discussed. See J. Inst. Metals 1933, 53, 474.—J. H. W. Adhesives lor Metal Foil and the Technique of Glueing It. Fritz Old (Gelatine, Leim, Klebstoffe, 1934, 2, 195-199; C. Abs., 1934, 28, 7437).— The adhesives used are discussed. The results depend as much on the technique of glueing as on the variety of glue.— S. G.

XXIII—BIBLIOGRAPHY

(Publications marked * may be consulted in the Library.) Chase, Herbert. Die-Castings: Their Design, Composition, Application, Specification, Testing, and Finishing. 1934. L o n d o n : Chapm an and Hall, Ltd. (21«. 6d.) *Copper Development Association. Sheet Copper Work for Building. A Practical Handbook. 4to. Pp. 69, with 58 illustrations. [1934.] London: The Association, Thames House, Millbank, S.W .l. »Friedmann, Werner. Bestimmung der Biegewechselfestigkeit von Drähten. Bau einer entsprechenden Materialprüfmaschine. (Mitteilungen des Wöhler-Instituts, Braunschweig. Heft 22.) Demy Svo. Pp. 93 with 60 illustrations. 1934. Berlin: NEM-Verlag G.m.b.H. (R.m! 3.60.) 38 Metallurgical A bstra cts Vol. 2

G.W .B. Electric Furnaces, Ltd. Quenching Oils. London : G.W.B. Electric Furnaces, Ltd., Elecfurn Works, North Road. Gil, Emilio Jimeno. Corrosión metálica. Pp. 42. 1934. Barcelona : Academia de Ciencias y Artes. Guérillot, ---- . Protection des métaux contre la corrosion. Pp. 200. 1934. Paris : J.-B. Baillière et fils. (25 francs.) Hartshome, N. H., and A. Stuart. Crystals and the Polarizing Microscope : A Handbook for Chemists and Others. With a Foreword by G. T. Morgan. Pp. viii + 272. 1934. London : Edward Arnold. (10s.) * Jacquet, A., et D. Tombeck. Éléments de Métallurgie. (Bibliothèque de l’Enseignement Technique.) Troisième édition, rédigée conformément aux nouveaux programmes de l’enseignement technique. Technologie des Marchandaises, 2e. année, 3e. partie. Demy 8vo. Pp. viii -f 363, with 222 illustrations. 1934. Paris : Dunod. (Br., 17 francs ; cart., 19 francs ; relié, 22 francs.) *Küchler, E. Föppl, 0. Untersuchungen an scheibenförmigen Resonanz- Drehschwingungsdämpfern bei hohem Schwingungszahlen. V o n E . K iic h le r. Pp. 1-77, with 46 illustrations. Das Dämpfungsmass bei Schwingungen. Von 0. Föppl. Pp. 79-84. (Mitteilungen des Wöhler- Instituts, Braunschweig. Heft 23.) Denn' Svo. 1934. Berlin : NEM-Verlag G.m.b.H. (R.M. 3.60.) Landgraf, Otto. Beitrüge zur Kenntnis des Systems Eisen-Wolfram. (F o r schungsarbeiten über Metallkunde und Röntgenmetallographie. Herausgegeben von Maximilian Frhr. v. Schwarz. Folge 12.) Med. 8vo. Pp. 46, with 22 illustrations. München und Leipzig : Fritz u. Joseph Voglrieder. (R.M. 3.) *Linde A ir Products Company. Precautions and Safe Practices in the Storage, Care, and Handling of Oxy-Acetylene Welding and Cutting Equipment. Pp. 24. 1934. New' York : The Company, 205 E. 42nd St. (Gratis.) Losana, Luigi. Lezioni di metallurgia. Pp. 350. 1934. Torino : A. Viretto. *Masing, G. Herausgegeben von. Handbuch der Metallphysik. B a n d I : Der Metallische Zustand der Materie. Erster Teil. Ciüeraufbau metallischer Systeme. Von U. Dehlinger. Grundlagen des metallischen Zustandes. Physikalische Eigenschaften der Metalle. Von G. Borelius. Sup. Roy. Svo. Pp. xiii -f- 520, with 213 illustrations. 1935. Leipzig : Akademische Verlagsgesellschaft m.b.H. (R.M. 46; Lw., R.M. 47.60; subscription price (4 Bände), R.M. 39.10 ; Lw., R.M. 40.45.) Rossié, H. Die Bedeutung der Zeitstudien für die Durchforschung und Wirt schaftlichkeit des Wassergasschiceissvorqanges. P p . 63. W ü rz b u rg : Triltsch Verlag. (M. 3.) *Royal Mint. Sixty-Fourth Annual Report of the Deputy Master and Comp troller of the Royal M int, 1S33. Roy. Svo. Pp. iv -(- 163, with 4 plates. 1934. London : H.M. Stationery Office. (3s. 6d. not.) [Tn the report of the Chief Assayer, the “ 720 ” silver-copper alloy, and the effec t of bismuth in coinage bronze are dealt with.] Sautner, Karl. Beitrag zur Kenntnis des Systems Kupfer-Silizium. (F o r schungsarbeiten über Metallkunde und Röntgenmctallographie. Herausgegeben von Maximilian Frhr. v. Schwarz. Folge 9.) Med. 8vo. Pp. 31, with 4 illustrations. München und Leipzig : Fritz u. Joseph Voglrieder. (R.M . 3.) v. Schwarz, (Frhr.) Maximilian. Röntgenschattenbilder von metallischen Werkstücken und ihre densographische Auswertung. (Forschungsarbeiten über Metallkunde und Röntgenmetallographie. Herausgegeben von 1935 X X III.—Bibliography 39

.Maximilian Frhr. v. Schwarz. Folge 8.) Med. 8vo. Pp. 35, with 23 illustrations. München und Leipzig: Fritz u. Joseph Voglrieder. (R .M . 3.) v. Schwarz, (Frhr.) Maximilian, F. Goldmann, und H. Fromm. Emaiüer- und Verzinkungsfehler. (Forschungsarbeiten über Metallkunde und Rönt genmetallographie. Herausgogeben von Maximilian Frhr. v. Schwarz. Folge 11.) Med. 8vo. Pp. 64, with 21 illustrations. München und Leipzig: Fritz u. Joseph Voglrieder. (K.M. 4.50.) ♦Schwinning, W . Konstruktion und Werkstoff der Geschützrohre und Gewehr läufe. D e m y8 v o . P p . v ii + 167, w ith 117 illu s tra tio n s . 1934. B e r lin : VDI-Verlag G.m.b.H. (Br., R.M. 15; VDI-Mitgl., R.M. 13.50.) ♦Science Library, London. Aluminium-Nickel-Iron Alloys. (Bibliographical Series, No. 108.) [Mimeographed.] Fcap. Pp. 1. 1933. London: The Science Library. ♦Science Library, London. Glossaries of Technical Definitions (Exclusive of Botany and Zoology.) (Bibliographical Series, No. 114.) [Mimeo graphed.] Fcap. Pp. 3. 1933. London : The Science Library. ♦Science Library, London. Impact Testing in General. (Bibliographical Series, No. 106.) [Mimeographed.] Fcap. Pp. 2. 1933. London; The Science Library. ♦Science Library, London. Short Lists of Text-Books on Structure of the Atom ; Isotopes; Radioactivity; Catalysis; Colloids; Complex Ions anil Co-ordination Compounds ; Theories of Valency; The Periodic System of the Elements; Allotropy; Intermetallic Compounds; The Rare Earths; Crystal Structure; Corrosion and Passivity and Protection of Metals. (Bibliographical Series, No. 45.) [Mimeographed.] Fcap. Pp. 7. 1932; re-issue 1934. London : The Scicnco Library. ♦Science Library, London. Spring Making. (Bibliographical Scries, No. 125.) [Mimeographed.] Fcap. Pp. 7. 1934. London: The Science L ib r a ry . ♦Science Library, London. Strength and Testing of Metals. (Bibliographical Series, No. 62.) [Mimeographed.] Fcap. Pp. 4. 1932; reprinted. London : The Science Library. ♦Siebei, E., und E . Kopf. Beanspruchung in gelochten Platten. (Forschungs heft 369.) 8J in. x 11J in. Pp. 22, with 72 illustrations. 1934. Berlin : VDI-Verlag G.m.b.H. (R.M. 5; VDI-Mitgl., R.M. 4.50.) ♦Society oE Public Analysts and Other Analytical Chemists. Compiled for. Bibliography of the More Important Heavy Metals Occurring in Food and Biological Material for the Years 1921 to 1933. inclusive. 8vo. Pp. 30. 1934. London: The Analyst, 85 Eccleston Sq., S.W .l. (Members, 2«.; non-members, 3s.) Sosman, Robert B., and Olaf Anderson. Large-Scale Phase Equilibrium Diagrams. Kearney, N .J.: Research Laboratory, U.S. Steel Cor poration. ($2.00.) ♦Thum, A., und F. Wunderlich. Dauerbiegefestigkeit von Konstruktionsleiten an Einspannungen, Nabensitzen und ähnlichem Kraftangriffstellen. (Mitteilungen der Materialprüfungsanstalt an der Technischen Hoch schule Darmstadt. Herausgegeben von A. Thum.) Med. 8vo. Pp. viii + 82, with 94 illustrations. 1934. Berlin : VDI.-Verlag G.m.b.H. (Br-, R.M. 7.50; VDI.-Mitgl., R.M. 6.75.) ♦Wendeborn, Helmut B. Saugzug-Sintern und -Rösten. Grundlagen und Anwendung der Saugzugverblaseverfahren. 6 J X 6} in. Pp. viii -{- 116, with 34 illustrations. 1934. Berlin: VDI.-Verlag G.m.b.H. (Br., R.M. 10; VDI.-Mitgl., R.M. 9.) 40 Metallurgical Abstracts Vol. 2,1935

*WiecheIl, Heinz Günther. Uber die Veredlmigsjühiqkcit der Gusslegierungen Alum inium -f 9% MgZnt nach ihren mechanischen und Korrosions- Eigenschaften. (Forschungsarbeiten über Metallkunde und Röntgen- metallographie. Herausgegeben von Maximilian Frhr. v. Schwarz. Folge 13.) Med. Svo. Pp. 52, with 30 illustrations. [1933.] München und Leipzig : Fritz u. Joseph Voglrieder. (R.M. 3.) W orden, E . C. United States Chemical Patents Index, 1915-1924. V olum e V.— Index of Subjects S, T, U, V, W, X, Y, and Z. P p . 1091. 1934. New York : Chemical Catalog Co., Inc. ($25.)

XXIV.—BOOK REVIEW S

Pattern-Making. By James Ritchey. Revised by Walter W. Monroe, Charles Win, Becse, and Philip Ray Hall. Med. Svo. Pp. vi + 237, with 359 illustrations. 1933. Chicago, 111.: American Technical Society. ( S I . 75.) The art of pattern-making has developed greatly in reccut years, just as other branches of the mcchanical Industries have expanded, and the work of the pattern-maker is one of the first fitcp3 towards the production of the complete machine. The complexity of the patterns lias increased in proportion as modem machines have become more complicated, calling for even greater skill on the part of the pattern-maker and the wider acquaintance with the various foundry methods which have their effect on pattcrn-construction. Duplication and mass production have led to the wider use of metal master patterns, and moulding machines for repetition work l-ave still further enlarged the field of pattern-making. The aim of this book is to cover fully the subject of pattern-making, giving the tools and equip ment necessary, the design details of simple and complicated patterns for typical cases, use of green- and dry-sand cores, and finally the construction and design of a typical moulding machine with details as to the manner in which tho castings are designed to suit this maehlnc. The authors achieve their aim and cover the subject adequately In a concise manner. The book is well written, being characterized by simplicity of language and clarity of expression. Not only will it prove of direct value to practical pattern-makers, but it contains much that is of interest to all connected with foundries who seek to enlarge their knowledge of this important branch cf foundry work. It is particularly recommended for foundry students and appren tices. Tho book is well printed and profusely illustrated with useful diagrams and photographs. —J. E. N ewso.v. Test-Book of Mechanical Engineering. [Notified in Arm y Orders for August, 1934.] Demy8vo. Pp. xv + 690, with 135 illustrations. 1934. London: H.M. Stationery Office. (12s. net.) In view of the development of mechanization In the Army tho issue of this manual by the Army Council is of special interest. Whilst it is intended specifically to serve as an instruc tional text-book for Army use, it presents such a comprehensive survey of the subject that it should prove to be a valuable reference book for many young engineers outside the service. I t is divided into three main parts—workshop practicc, heat engines, and applications. In a work of this scope it is obviously impossible to devote much space to metallurgical con siderations, but a brief r(sum( is given of the commoner ferrous and non-ferrous materials used in the branches of mechanical engineering covered by the book. Very useful chapters include the principles and practice of hcat-treatmcnt, pattern-making, foundry work, and Joining of metals. . A broad outline of workshop practice in these subjects is given, and similar treatment of fitting shop, macliine shop, and Wood-working shop follows. This part is con cluded by an excellent chapter on workshop economics. The second part forms a very complete treatise on heat engines and deals very fully with the theory, care, and maintenance of internal combustion engines—gas, petrol, oil, ami licavy-oil Diesel types. The overhaul, repair, and testing of these engines receive naturally considerable detailed attention. Eight chapters are devoted to steam boilers and engines, their construction, auxiliaries, and working. The final part on transmission of power, lubrica tion, compressors, and pneumatic tools and refrigeration is equally thorough. The whole volume is profusely illustrated with line drawings, photographs, and diagrams, and is well Indexed. Following the index Is a bibliography arranged under tho chapter numbers, and the only criticism that one may offer in this respect is that the references given to metal lurgical text-books could have been rather more comprehensive. Considering the amount of information gathered toget her In this book the price is remarkably low, but the quality of the paper may account in part for the lower cost of production. —J. E. NEWSON. EXTRUDED

PRESSINGS. CHILL CAST BARS, AND

BROTHERS LIMITED Brass Rods, Stampings and Non-ferrous Ingot Metal Manufacturers. METAL WORKS: ROTTON PARK STREET, BIRMINGHAM 16 T tltfh tn t: Edgbajton (JÎ&0 {T lints).

T c h f r a t n s : ** MeJCtchnit, SirrH inxkam .” LONDON—-17 Victoria St., Westminster, S.W. 1. N E WCASTLH-ON-TVXE-90 l'ilZrim Street. MANCHESTER—5U-ï3ContExchange Bldgs. 4 I.EE DS—Prudential Buildings, Park Row, SMELTING W orks : W J ONES, LANCS.

Engineers, Foundry Managers You ought to Foremen or Artisans become a member of (he FOUNDRY TECHNICAL ASSOCIATION which will bring to you most useful and important data through its MONTHLY JOURNAL THE WORK AND its BI-MONTHLY CONFERENCES & RESEARCH OF ITS its ANNUAL CONGRESSES COMMITTEES You will be kept continually conversant with the progress realised in the sphere of your profession; You will benefit from the experience of your colleagues; You will gain lime by adding quickly to your experience.

Write for particulars to the ASSOCIATION TECHNIQUE DE FONDERIE 44, rite de Rennes, PAEIS (YI°) CONSOLIDATED TIN SMELTERS LIMITED PRINCES HOUSE 95, GRESHAM STREET, LONDON, E.C.2 Telephone, : Met. S946-8. Telegram : Ccnsmelter, ’Phone, London.

Proprietors of the following Brands of Tin : STRAITS E. S. COY., LTD., PENANG Works: PENANG, Straits Settlements. ENGLISH

M elianear (99-9% Guaranteed). Cornish LAMB M ellanear &

Penpoll (Common & Refined). FLAG

W orks ï BOOTLE, Near Liverpool.

1Buyers of A ll Classes of Tin Ores.

Sole Selling Agents : VIVIAN, YOUNGER & BOND, LIMITED 8, Basingliall Street, London, E.C.2 Telephone: MET. 2061. Telegrams: BOND, TELEW, LONDON.

P r i n t e d i n G r e a t B r i t a in b y R ic h a r d C l a y a n d S o n s , L t d ., B u k g a y , S u f f o l k .