Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

An investigation of ferromagnetic phases in manganese rich alloys Shirkoff, George Peter Alexandroff 1953

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1953_A7 S5 I5.pdf [ 3.59MB ]
Metadata
JSON: 831-1.0106434.json
JSON-LD: 831-1.0106434-ld.json
RDF/XML (Pretty): 831-1.0106434-rdf.xml
RDF/JSON: 831-1.0106434-rdf.json
Turtle: 831-1.0106434-turtle.txt
N-Triples: 831-1.0106434-rdf-ntriples.txt
Original Record: 831-1.0106434-source.json
Full Text
831-1.0106434-fulltext.txt
Citation
831-1.0106434.ris

Full Text

AN INVESTIGATION OF FERROMAGNETIC PHASES IN MANGANESE RICH ALLOYS by GEORGE PETER ALEXANDROFF-SHIROKDFF A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR TTTR DEGREE OF MASTER OF APPLIED SCIENCE in the Department of Mining and Metallurgy We accept this thesis as conforming to the standard required from candidates for the degree of MASTER &F APPLIED SCIENCE. Members of the Department of Mining and Metallurgy THE UNIVERSITY OF BRITISH COLUMBIA April, 1953. ABSTRACT The investigation of ferromagnetic phases in the manganese indium, manganese antimony carbon and the manganese magnesium alloy systems was conducted mainly by means of x-ray diffraction methods and where ever permissible by metal1ographic examination. The manganese indium and the manganese antimony carbon alloys were prepared in the vacuum melting furnace in an argon atmosphere, while the manganese magnesium alloys were prepared by sinter-compact techniques. Heat treatments were carried out in tube furnaces in an argon atmosphere. Ferromagnetic phases were not found in either the manganese indium or the manganese magnesium systems. The presence of the compound Mn^In in the manganese indium system was established. The compound has a gamma brass structure with a lattice parameter *q ~ 9.413 A°. In the manganese magnesium system only solid solutions of manganese and magnesium were found to exist. The manganese solid solution was not magnetic suggesting that the addition of magnesium atoms did not give the degree of separation of manganese atoms for favourable ferromagnetic conditions. It was concluded also that the manganese atoms in the compound Mn^In were not sufficiently separated for favourable ferromagnetic conditions. The ferromagnetic phase found in the manganese antimony carbon alloys was attributed to the compound Mn^ Sb. This suggested that elements in Group Vb which are strongly electronegative to manganese and form stable binary compounds, do not tend to form ternary compounds containing carbon. •• _ ; ACKNCWLELXJEMEHT The author is grateful to the staff of the Department of Mining and Metallurgy for their consideration and interest; and to Associate Professor W. M. Armstrong for his helpful criticism and generous encouragement during the investigation. The author gratefully acknowledges the aid rendered him by Mr. R. Butters who assisted in numerous technical problems that arose in this investigation. The author gratefully acknowledges the Consolidated Mining and Smelting Company and the Electro-Manganese Corporation of America for donating electrolytic manganese and indium metals. This work was made possible by a grant from the Defense Research Board. TABLE OF CONTENTS I INTRODUCTION 1 II PREVIOUS WORK . . 8 III PROCEDURE 13 IT RESULTS . 20 (a) Manganese Indium Alloys 20 (b) Manganese Antimony Carbon Alloys 23 (c) Manganese Magnesium Alloys . . . 23 V DISCUSSION 2 8 (a) Manganese Indium Alloys 28 (b) Manganese Antimony Carbon Alloys 29 (c) Manganese Magnesium Alloys 30 VI CONCLUSION 35 VII APPENDIX 36 (a) Metals used in this Investigation 36 (b) Vacuum Melting Equipment 37 (c) Tube Furnace and Bomb 40 VIII BIBLIOGRAPHY ^ 2 ILLUSTRATIONS page Fig. I- The Saturation Magnetisation Per Atom of Ferromagnetic Alloys as a Function of the Total Number of Electrons per Atom in the 3d and ks Bands It-Fig. II- Energy of Magnetisation as a Function of Radius of Atom/Radius of Ion 7 Fig. I l l - The Equilibrium Diagrams of the Systems Manganese . Indium as reported by Goeddel and Yost, and Zwicker . . 9 Fig. IV- A Portion of the Equilibrium Diagram of the System Manganese Magnesium . 12 Fig. V- Metallographic Photograph of 25 Per Gent Indium Alloy Anneated at 800°C for l68 hours and Water Quenched . . 21 Fig. VI- Metallographic Photograph of 33 Per Cent Indium Alloy Anneated at 150OC for 168 hours and Water Quenched . . 22 Fig. VII- X-ray Photograph of 25 Per Cent Indium Alloy 22 Fig. VIII X-ray Photograph of 25 Per Cent Magnesium Alloy . . . . 27 Fig. IX- Diagrammatic View of Vacuum Melting Unit 38 Fig. X- Rear View of Vacuum Melting Furnace and Evacuation Unit 39 Fig. XI- Diagrammatic View of Tube Furnace kl T A B L E S page Table I variations of Lattice Parameters in Manganese Indium Carbon and Manganese Tin Carbon Alloys . . 10 Table II Results of Manganese Antimony Carbon Alloys . . . 2k Table III Results of Manganese Magnesium Alloys 26 AN INVESTIGATION OF FERROMAGNETIC PHASES IN MANGANESE RICH ALLOYS. I - INTRODUCTION A l l materials may be classified either as ferromagnetic, paramagnetic or diamagnetic. Materials which are strongly magnetic in the presence of an external magnetic fie l d and remain magnetic when the field is removed are called ferromagnetics. Paramagnetic materials, which are more numerous than ferromagnetics, are weakly magnetic and only exhibit magnetization in the presence of an external magnetic field. In diamagnetic substances the magnetization is directed oppositely to the external magnetic field, they are therefore repelled from the poles of a magnet. Many of the metals and most of the nonmetals are diamagnetic. The source of ferromagnetism is said to be the magnetic moment associated with the spin of the electron. A current flowing through a wire gives rise to a magnetic field. Since the motion of the electrons produces this flow of current, a rotational motion of an electron wil l give rise to a magnetic moment. The electron exhibits two rotational motions in a metal: an orbital motion around the nucleus and a spin motion about its own axis. The magnetic moment produced by the orbital motion is small compared to the magnetic moment produced by the spin motion and may be ignored. A ferromagnetic material has more magnetic moments aligned 2 In one direction than In another, that i s , there are more electrons spinning in one direction than in another. This unbalanced alignment of magnetic moments is only possible i f the energy states are partially f i l l e d . According to Heiseriberg1 the energy that aligns the electrons is a result of the electrostatic exchange interaction between the electrons and the atoms. When the density of states is low, that i s , when the energy difference between the levels is large, there is insufficient energy to align the electrons in these states. The outer valency electrons are usually in partially f i l l e d states but their density of states is low and therefore they do not contribute to ferromagnetism. According to the Pauli exclusion principle, no two electrons, in a given atom can be in the same quantum state; therefore, each pair of electrons in the same level must have opposite spins and their magnetic moments cancel. The principle applies to a l l states below the 3d in the elements below chromium in the periodic table. In the transition metals, there are partially f i l l e d 3d states in the iron group and partially f i l l e d k$ states in the rare earths. These states have a high density of states and do not require large energies to align the electrons. These metals therefore have a favourable electronic structure for ferromagnetism. The relative saturation magnetization of these magnetic trans-ition elements may be compared to the number of electrons in the 3d band of states at absolute zero. Since the 3d and the he bands of states overlap, i t is not necessary to have an integral number in each band, but the total number of electrons in the bands must remain the same. The 3d band has five quantised energy states. This band may be divided into two half bands, one containing electrons with positive spin the other 3 containing electrons with negative spin. The maximum saturation occurs when the available electrons from the 3d and ks bands f i l l the f i r s t half band leaving the remainder of electrons for the second half band. The saturation per atom is proportional to the difference in number of electrons between the two half bands. The experimental value for the saturation magnetization for nickel is 0.6 Bohr magnetons ; where the Bohr magneton is defined as the quantum mechanical unit of magnetic moment contributed by a single electron. Therefore there must be 0.6 electrons more of one type of spin than another type of spin in the 3d band. Furthermore, in nickel there are a total of 10 electrons in the 3d plus ks band of states; therefore, there must be a total number of 0.6 electrons in the ks band of states. Assuming this value of 0.6 electrons in the ks band of states to be constant, a plot of saturation magnetization against the number of electrons in the 3d and 4s bands of the elements from chromium to copper would have a linear relationship as shown in Figure 1. However, the experimental curve does not follow this linear relationship and passes through a maximum at 2.5 Bohr magnetons saturation 2 magnetization. Coles explains the difference of 0.35 Bohr magnetons between the theoretical saturation magnetization and the experimental saturation magnetization for iron to be due to the lack of exchange interaction to produce complete saturation at absolute zero. 3 Pauling advanced a theory which is essentially an interpret-ation of the experimental curve. The theory states that part of the electrons in the 3d band take part in cohesion as do those of the ks, while the remainder contribute to ferromagnetism. Pauling divides the 3d band in two, the cohesion band and the atomic band responsible for ferro-Figure I. - The Saturation Magnetization Per Atom of Ferromagnetic Alloys as a Function of the Total Number of Electrons Per Atom in the 3d and ka Bands. 5 magnetism. Starting with potassium and moving from l e f t to right in the periodic table, the 3d band contributes one additional electron for every successive element until vanadium is reached with five contributory electrons. The next element chromium contributes 5.3 electrons and the cohesion band is considered to be f i l l e d . Any further available electrons, are these contained in the atomic band. According to Paulings theory the half band discussed previously can only hold 2.4 electrons per atom. Following the experimental curve and starting with nickel, the available electrons for the atomic band decreases and the magnetization increases until there are only 2.4 electrons available. This point corresponds to 8.2 total electrons available from the 3d and the 4s bands. Beyond this point, magnetization decreases as the total available electrons in the atomic band decrease. In the iron group of transition metals only iron, cobalt and nickel and in the rare earths gadolinium are ferromagnetic. Although pure manganese is not ferromagnetic, some manganese alloys are ferromag-k netic. Slater using Heisenbergs theory relates this lack of magnetism in some transition metals to the large radius of the 3& shell compared to the small interatomic distance of the element, fieisenberg states that i f two atoms are at a certain distance apart, each atom having a magnetic moment of one Bohr magneton due to the spin moment of one electron, a force of interaction exists between them, in addition to the electrostatic 5 and magnetic forces. Bethe's calculations show that as the two atoms are brought close together these forces cause the electron spins in the two atoms to become parallel (positive interaction). On decreasing the distance the parallel spin moments become more stable until at a certain distance the forces diminish and become zero. AB the distance is decreas-ed further the spins set themselves antiparallel (negative interaction). Slater and others show that as we pass backward along the periodic table from copper through nickel to chromium, the radius of the 3d shell increases, but the interatomic distances remain fairly constant. Under these conditions the 3d shell of adjacent atoms for manganese and chromium will be very close resulting in a negative exchange. The curve relating the exchange energy of magnetization to the ratio of the atomic separation and the diameter of the unfilled 3d shell is shown in Figure 2. The position of the various elements on the curve are not to~be taken as definite as the interaction varies with structure and the number of near-est neighbours. The condition of ferromagnetism therefore involves the lattice spacing of the metals and is thus structure sensitive. Manganese has an electronic structure favourable for ferromagnetism but is not ferromagnetic. By considering the radius of the 3d shell for the element as constant, and i f the atomic distance is increased by compound or alloy formation, man-ganese may exhibit ferromagnetism. The program conducted by Morgan^ of which this investigation was a part was concentrated upon the manganese rich alloys because of the slight deviation of the radii ratio (l.^7) for manganese from the critical ratio (l.^) suggested by Slater. Morgan proposes that the most suitable form of manganese for ferromagnetism is the high temperature tetragonal gamma phase. This investigation was therefore carried out at high temp-eratures in an attempt to derive ferromagnetic phases from the gamma manganese structure. The investigation of the system manganese-indium and manganese-7 antimony-carbon was proposed by Morgan in view of the m a y - i r a n m occuring in the liquidus of the manganese-indium system indicating a compound formation and the discovery of ferromagnetism in ternary systems containing manganese, carbon and another metallic element. The most magnetic alloys found in the ternary systems were those containing manganese aluminum carbon and manganese zinc carbon. The manganese a / H i m - t i m m carbon alloys were invest-igated by Morgan while the manganese zinc carbon alloys were investigated 7 by Butters . The relative position of magnesium in the periodic table to aluminum and zinc; furthermore, the size of the magnesium atom being approximately that of aluminum suggests that ferromagnetism may exist in the manganese magnesium carbon alloys. However, in view of the contro-versial reports on the manganese magnesium system, the investigation of this binary system was undertaken. C O B A L T .0 (Z It 2 Z o M A N G A N E S E R A T O M I C S E P A R A T I O N r D I A M E T E R O F U N F I L L E D S H E L L Figure II. - Energy of Magnetization as a Func-bion of Badiue of a Atom/Radius of Ion (After Slater). 8 II PREVIOUS WORK The existence of ferromagnetism in the manganese indium system between 3 and 50 weight per cent manganese was reported by Goeddel and 8 Yost . The preparation of the alloys was carried out in alundum crucibles with graphite sheaths under an argon atmosphere. Qualitative magnetic measurements were made with a small Alnico magnet. The authors suggested that ferromagnetism may be due to a single phase, the compound Mn2In. Zwicker^ in an earlier investigation of this alloy system reports the existence of the compound Mn^In. The equilibrium diagrams reported by these investigators are shown in Figure 3* Further work was conducted in this system by Eppelsheimer and Barnes10 reported that alloys from 10 to 50 weight per cent indium were found to be non magnetic. These observations contradict Goeddel*s and Yost's observations at 50 weight per cent manganese corresponding to the compound Mn2In. 6 Piercy , in his investigation of the manganese aluminum system found the existence of a face centred cubic ferromagnetic phase. The investigation of this phase was conducted by Morgan and was found to have a composition Mn^2Al23C15 * ^ e c a r ^ o n absorption was attributed to the use of carbon crucibles. As a result, Morgan conducted a series of investigations of the ternary MnXC systems; where X represents a metallic element which has both a positive size factor with respect to manganese and a high positive valence. High temperature ferromagnetic phases were found to exist in three systems; those containing aluminum, * The composition of many alloys is shown in this form, they are not - considered to be intermediate compounds. 9 A /4.0O / • o o I30Q - / 3 0 a / 2 0 0 1 i z o a // OO I OOO i net-r UOO / o o o 9 0 0 / S o o 8 0 0 • j -S-5. . . . 0 ^ 6 0 0 j 1 \ \ \ \ \ 1 \ I \ S 9 o " 7 0 0 • ' 3 •4. 6 O O ^ j 1 1 1 \ I T o o I - 0 0 - 0 300 z o o - ; , I tS7. S 0 2 0 0 /.JO - I 1 1 1 1 1 1 1 . 1 I • 1 • I I I , . 1 0 • » -?o J o feo v o 5 0 y o / o o < 3 /O r o j o 4 0 . 5 0 ( 6<? 7 0 y o y o / o o -4 /.aw. ^ Csi<jtvm Figure III. - The Equilibrium Diagrams of the System Manganese Indium as reported by Goeddel and Yost, and Zwicker. 10 Indium and tin. The phase was reported as face centred, cubic in a l l systems and showed a superlattice formation In the manganese aluminum carbon and the manganese tin carbon alloys. The superlattice was report-ed to exist over a vide range of composition. The magnetic phase vas concluded to be based on the composition (MnX)hC. In the temperature range of 1050°C - 6O0°C. In both the manganese Indium carbon and manganese tin carbon systems, two face centred cubic lattices were report-ed to exist on homogenisation, while the cast alloys exhibited only one face-centred cubic lattice. In the alloy M % 2 S n 2 1 C l 6 M°rsan reports two face centred cubic lattices exhibiting superlattices. In the man-ganese aluminum carbon system only the alloy Mn^ ^ 7^21C15 s i l o w e ^ t " 0 face centred cubic structures on furnace cooling from 900°C. The variation of lattice parameter of the two face centred cubic cells is shown in Table I; the parameters increases with increasing X. TABLE I. Lattice System • Composition Range in atomic # Variation of lattice parameter. small MnlnC 1 - 6 3.87 - 3.871 large 3.898 - 3.9^0 small MnSnC 2 - 1 0 3.882 large 3.900 - 3.981 The investigation of the manganese magnesium alloys of high 11 magnesium content was conducted by Schmidt . The existence of alpha manganese in equilibrium with alpha magnesium from the solidus to room temperature in the range of 0.1 to 3 per cent** manganese was reported. The investigated portion of the equilibrium diagram is shown in Figure k. In the diagram the Greek letter alpha represents alpha magnesium and the letter beta represents alpha manganese. The solid lines of both the 12 liquidus and the solidus are those reported by Grogan and Baughton, 13 while the clotted lines are those reported by Schneider and Scholder . The investigation was repeated and the same equilibrium conditions were found to exist by the following: Siebel*, Grogan and Maughton, Timer , 16 Schneider and Scholder. Bakhmeter and Golovchiner on the basis of Laue diffraction patterns report the second phase that separates from the liquidus as beta manganese. A compound Mg^ Mn was reported by Sawamoto 13 and Janaki ; however Schneider and Scholder in their investigation . found no such compound and report the existence of only beta manganese in equilibrium with alpha magnesium on quenching from 100°C in the range of 50 per cent manganese. A l l per centages refer to atomic per cent unless otherwise stated. 'C Atomic Percentage Manganese °F / 2 650 750 650 550 \' L .—-"•"^  Q 650" + L 2.1(2.45. P + -651- -L a ~7\ a * P . -1500] 1300' 1100 Mg I 2 3 4 5 6 Weight Percentage Manganese Figure IV - A Portion of the Equilibrium Diagram of the System Manganese Magnesium. I l l PROCEDURE The purity and form of the metals used in making the alloys are listed in Appendix 1. Since manganese nitride (Mn^ N) is ferromagnetic, the manganese indium and the manganese antimony carbon alloys were melted in a vacuum melting furnace (described in Appendix 2.) in the presence of a reducing agent and in an argon atmosphere. A specific changing method was adopted in the case of the manganese indium alloys as preliminary melting procedures shoved that indium metal in the molten state vets the surface of the alundum crucibles. By placing the metal on the bottom of the crucible the least area of surface contact was obtained and this reduced the possibility of loosing the metal on the walls of the crucible. The general melting procedure was as follows: the furnace *** was evacuated to a pressure of 50 microns of mercury while the charge was heated with the hope of degasifiing the interior of the unit and the surface of the metal particles. Hydrogen was used as a reducing agent in the manganese indium alloys. The furnace was flushed with the gas after evacuation and the charge melted under a pressure of 6 ins. of mercury. The furnace was then flushed several times with argon and the alloy remelted under a half an atmosphere of argon, and cast. The loss of manganese due to i t s vapour pressure at high temperatures did not permit melting the alloys in vacuo. *** A l l pressures are absolute pressures. As carbon is a good reducing agent, the presence of carbon in the manganese antimony carbon alloys did not necessitate the use of hydrogen as a reducing agent. The constituents were mixed prior to be-ing placed in the crucible; however, a fairly thick layer of crushed manganese had to be placed on top of the remaining charge in the crucible to prevent the light particles of carbon from escaping daring evacuation. The melting procedure was otherwise the same as for the manganese indium alloys except for the reducing step just discussed. o Magnesium has a vapour pressure of one atmosphere at 1150 C. This temperature is approximately 100°C below the melting point of manganese and therefore necessitated a revision of the melting method for the manganese magnesium alloys. A number of attempts were made to employ the vacuum furnace melting method using approximately l£ atmosphere' of argon; however, the loss of magnesium through distillation was in-tolerable. A closed bomb method was employed. Manganese of -60 mesh and the finest available magnesium was used. The powders were fi r s t thoroughly mixed and then pressed in a die under a pressure of 55,000 psi using a hydraulic tensile testing unit. The compact was suspended in 80 per cent magnesium oxide and 20 per cent magnesium in a bomb described in Appendix. The only other suspension medium that could have been used and avoided'considerable oxidation of magnesium is calcium oxide; however, calcium fuses very readily with magnesium producing a low temperature entectic which would be undesirable. After the ends of the bomb were welded the unit was heated in a tube furnace (shown in the Appendix III) at 700°C(50°C above melting point of magnesium) for 12 hours and then at 1000°C for 72 hours. Preliminary trials with various sintering temperatures and times showed that this procedure was the most satisfactory. However, these preliminary alloys were found to contain some magnesium oxide; as a result, the procedure was modified. A pin hole was left in the top of the bomb after the edges were welded. The bomb was then placed in the tube furnace and evacuated over a period of o an hour at 200 C. The furnace was then f i l l e d with argon to a pressure above atmospheric with the hope the gas would penetrate the bomb through the pinhole and provide an argon atmosphere. On removing the bomb from the furnace the pinhole was immediately welded. This method of pre-paration eliminate the oxidation of magnesium to some degree; however, there was s t i l l evidence of the oxide existing. The tube furnaces mentioned were used for homogenizing and aunealing for a l l the systems investigated. Homogenization and annealing was conducted under an argon atmosphere. On completing the t r i a l melting procedures of the manganese indium alloys, ten alloys containing 5, 10, 15, 25 (Mn^In), 3*t- (M^In), kO, 50, 75 and 90 atomic per cent indium were melted and c h i l l cast. As these alloys were found to be non magnetic i t was thought that the rate of cooling may be contributing to magnetism. A second group of similar alloy was melted and slow cooled and the results were negative. In view of these results high temperature investigations were carried out. The cast alloys containing 5, 10, 15, 25 per cent indium were homogenized at 800°C for 168 hours in the region where Goeddel reports an unknown solid solution. The homogenized alloys were then annealed o at 500 C for 12 hours to investigate the region where Zwicker reports the existence of alpha manganese in equilibrium with the compound Mn^In. The cast 3^ per cent indium alloy corresponding to the compound Mn2In o was homogenized at 150 C for loo hours. The alloys after homogenizing 16 and annealing were polished, etched with two per cent nital and examined metallographlcal1y. X-ray diffraction analysis was made on the 10, 25 (Mn^In) per cent indium alloys and the annealing scale found coating the alloys after heat treatment. A Philips X-ray diffraction apparatus was used for the X-ray vork. The photographs were produced with a 14.32 cms diameter Ievins-Strausmanis camera. Films were exposed for three hours under iron radiation without a f i l t e r at 30 KV and 13 milliamps. These long ex-posures were found to be unnecessary and in the investigation of the manganese antimony carbon and manganese magnesium systems the exposures were reduced to 2 hours at kO KV and 10 milliamps. In the investigation of the manganese antimony carbon alloys the f i r s t group of alloys melted contained 20 per cent carbon and vary-ing amounts of manganese, because the chemical composition of the ferro-magnetic phase found in the manganese aluminum carbon system was in this range. The compositions of these manganese antimony carbon alloys were as follows: ^ 2 0 C20 ^ 0 8blD C20 • *T3 S b 7 C20 Ma^ Sb5 l f a 77.6 S b2.4 C20 M n 7 9 3 0 c 20 Except for the alloy containing 10 per cent antimony, the alloys were c h i l l cast. The 10 per cent antimony alloy was slow cooled to investigate the effect of cooling rate on ferromagnetism. In addition 17 to this group of alloys, two alloys of different carbon composition were melted, one containing manganese to antimony in the ratio of 2:1 (MngSb), the other with a similar manganese to antimony ratio except for an ad-ditional four atomic per cent carbon (Mn^kSb^Ck)* The purpose in making these alloys was to investigate the change in crystal structure of the binary compound Mn^ Sb and to investigate changes in magnetization on addition of carbon atoms. Preliminary heat treatments showed that the alloys decomposed in air in approximately two hours on quenching from high temperatures. The alloys decomposed to a brown powder and the X-ray diffraction pattern was complex with the back reflection lines very broad. Morgan reports that the X-ray diffraction patterns of the decomposed manganese indium carbon alloys contained a face centred cubic phase; thisJphase, however, was not found in the X-ray diffraction patterns of the decomposed manganese antimony carbon alloys. As a result of this decomposition the alloys were slow cooled on homogenization. The following alloys were homogenized at 900°C for 72 hours and furnace cooled. ltoT5 S b7 c20 Mn?5 Sb5 Mng Sb m6k S b32 CU The homogenized alloy Mo^Sb^CgQ was annealed at 900°C for UQ hours and water quenched. X-ray diffraction examination was conducted on a l l homogenized alloys, annealed alloys and also the following c h i l l cast alloys: *°TI.6 Sl02.k c20 Mngk Sb 3 2 CgQ Mn2 Sb 18 Qualitative magnetic measurements were made on a l l the alloys with a small Alnico magnet* Furthermore relative quantitative magnetic measurement were made on the homogenized alloys Mng^Sb^CjaQ with an electric magnet described in the Appendix. In the investigation of the manganese magnesium system, seven alloys were prepared using the revised bomb technique. The alloys cont-ained 10, 15, 20, 25 (Mn^), 29, 33 (MngMg) and 50(MnMg) per cent magnesium. As the loss of magnesium at 1000°C was found to be appreciable, the alloys were homogenized at 900°C for 72 hours and then annealed and water quenched from temperatures ranging from 500°C to 900°C. The anneal-ing vas carried out in steps. The alloys were slow cooled in steps of o 100 C from the homogenizing temperature and were heat treated at each temperature for 2h hours. On reaching the final annealing temperature, the alloys were heat treated for kQ hours and water quenched. The total time for heat treatment ranged from 72 to 192 hours per alloy. The alloys containing 25 and 33 per cent magnesium were homo-genized at 500°C and 600°C for 72 hours to a week to avoid retaining any high temperature phases. Mixtures of manganese and magnesium powders were heat treated in quartz tubes. Two manganese magnesium powders with composition Mn^ Mg were placed in quartz tubes. One of the tubes was evacuated and sealed, the other tube was left open. The powders in the tubes were then heat treated at 900°C for 2h hours in a nitrogen atmosphere and water quenched. The same quartz tube procedure was adopted In heat treating powders in an ammonia atmosphere for 20 hours. These powders were water quenched also. The composition of these powders was 5, 10, 15 and 25 (Mn^ Mg) per cent magnesium. Because of the physical properties of the compacts, metallo-graphic examinations were not permissable and the investigation was conducted using X-ray diffraction techniques. IV RESULTS (a) Manganese Indium Alloys. A l l the alToys prior and after heat treatment In the manganese Indium system were found to be non magnetic. Metallographic examination of the 3^  per cent indium alloy (Mn^In) revealed that the alloy was two phase, while the 25 per cent indium alloy (Mn^In) was found to be single phase. The annealing scale coating the alloys after heat treatment was found to be magnetic. X-ray diffraction studies of this scale showed two face centred cubic lattices. One of the face centred cubic structures was identified as manganous oxide, Indicating oxidation taking place in the annealing furnaces. The identification was carried out by comparing the * d* values of this structure with thoBe of the American Society for Testing Materials card system. As manganous oxide is not magnetic, the second face-centred cubic phase was contributing to magnetism. This second phase was not identified, but is believed to be a manganese nitride (MnkK). X-ray examination of the 10 per cent Indium alloy homogenized at 800°C showed a beta manganese structure, while that of the 25 per cent Indium alloy (Mn^In) revealed a gamma brass structure with a lattice parameter a Q a 9,kl3 A°. The gamma brass structure was identified by comparing the X-ray photograph with that of gamma brass. Figure V. - 25 Per cent Indium Alloy (Mn.In) Annealed at 800°C for 168 hours and Water Quenched. (Mag. 800X) 22 Figure VII. X-ray photograph of 25 Per Cent Indium Alloy (Mn?In); Annealed at 800°C for 168 hours and Water Quenched. The fi l m shows a Brass Type Structure. ( a Q = 9.413 A°) 23 (b) Manganese Antimony Carbon Alloys. Alloys in the 20 per cent carbon group of the manganese anti-mony carbon system and containing less than 7 per cent antimony melted very cleanly, while those alloys containing over 7 per cent antimony reacted very violently in the molten state. On slow cooling the 10 per cent antimony alloy, a crust of rejected carbon formed on the surface of the ingot. The results of the X-ray diffraction examinations of the alloys are shown in Table II. The compound Mn^b was identified by comparing the X-ray photo-graph with that of the alloy corresponding to Mn^ 3b. The structure of the manganese carbide phase mentioned in Table II was identified by comparing with the X-ray pattern of Mn^C^ obtained from the work conduct-ed by Morgan. Qualitative magnetic measurements showed that magnetism in-creased with antimony content and not with carbon content. The relative quantitative measurements of the alloys M n ^ S b ^ and Mn^b were identical. (c) Manganese Magnesium Alloys. X-ray diffraction examination of the alloys furnace cooled in the bombs and ranging in composition from 10 to 50 per cent magnesium shoved alpha manganese in the alloys below 33 per cent- manganese and alpha manganese In equilibrium vith alpha magnesium In alloys with compositions above 33 per cent magnesium. T A B L E II Alloy Type Compound Found Present. Per Cent Chill cast Mn Sb 2 65 35 M n7 5S b5C20 Homogenized Mn2Sb Mn 2 3C 6 95 5 Annealed » Mn^b Mn 2 3C 6 95 5 Homogenized Mn^b *23 c6 98 2 Chill cast and Homogenized MngSb 100 Mn^ Sb 2 Chill cast and Homogenized Mn Sb 2 100 The diffraction patterns of the alloys containing 25 and 33 per cent magnesium and furnace cooled from 900°C shoved alpha manganese i n equilibrium with a face centred cubic phase. Qualitative magnetic examination revealed that the alloy was fairly strongly magnetic. These results contradict the X-ray diffraction results for the alloys of the same composition and furnace cooled in the bombs. Diffraction results of the quenched alloys from temperatures o ° ranging between 500 C and 900 C and chemical compositions varying from 10 to 50 per cent magnesium are shown in Table III. X-ray photographs of the alloys heat treated in a nitrogen atmosphere shoved an ordered face centred cubic lattice. Qualitative magnetic measurements of these alloys did notgShow any variation i n magnetism. 1 rcos 8 cosSv The lattice parameters extrapolated to ^Vgin Q * —Q—) - 0 showed no noticable variation. The lattice parameter values for the alloys Mn^ ijMg^  and Mn^ Mg were a Q = 3.8629 A° respectively. The only difference between the two films was the presence of magnesium oxide lines i n the Mn^ Mg alloy. The omission of these magnesium oxide lines in the Mn^ M^g^  alloy i s probably due to the lev scattering factors of magnesium and oxygen. A l l films of alloys containing over 15 per cent magnesium shoved magnesium oxide lines. The film of the alloy containing 25 per eent magnesium and heat treated in vacuo in a sealed quartz tube shoved beta manganese i n equilibrium with alpha magnesium. Thermal analysis of the alloys containing 25 and 33 per cent magnesium shoved no arrest points; however, both alloys on visual examin-ation were found fused. Fusion was probably due to magnesium. T A B L E III. Composition Heat Treatment Crystal Structure Per cent Ferro-magnetism M n 90 M g 10 900°C Beta manganese F.C. Tetragonal 20 80 non magnetic 700°C F.C. Cubic 30 weak Mng5Mg15 800°C F.C. Cubic F.C. Tetragonal 20 80 weak Mn^ Mg 90G°C F.C. Tetragonal a Q = 3.775 A° c7a . -9773 100 non magnetic 700°C F.C. Cubic 70 Mn^ Mg 900°C F.C. Cubic a c = 3.851 A° 100 strong 8CO°C F.C. Cubic a Q r 3.8571 A° 100 strong 700°C F.C. Cubic ao 5 3.867 A° F.C. Cubic a Q = 3.832 A° MnO. 60 30 10 medium-strong 6CO°C F.C.C.(small) F.C.C.(large) MnO 30 60 10 medium-strong 500°C F.C.C. a Q a 3.846 A alpha manganese 50 50 medium MngMg 900°C F.C. Cubic 100 strong 8ooGc F.C. Cubic 8 ^ 3 3.857 A° 100 strong 700°C F.C.C.(small) F.C.C.(large) S 30 65 5 medium-strong MnMg 800°C F.C.C.(small) F.C.C.(large) . s _ -30 60 10 medium Mn^ Mg F.C. to 400°C A.C. to B.T. F.C. Cubic a n . 3.869 A° 100 strong 27 Figure VIII. - 25 Per cent Magnesium Alloy Heat treated and Water Quenched from 800°C. Shoving Typical Ordered Face Centred Cubic Lattice with presence of Magnesium Oxide Lines. 28 V. DISCUSSION (a) Manganese Indium Alloys* The metallographic examinations of the 33 per cent indium alloy (Mn2In) shoved the alloy to contain two phases and the 25 per cent indium alloy (Mn^In) shoved the alloy to be a single phase structure. These observations are In agreement with Zvickers investigation. Also in agree-ment vith Zvickers report are the results of the X-ray examination of the 10 per cent Indium homogenized alloy vhich shoved only a beta:, manganese structure and the 25 per cent indium alloy (Mn^In) which revealed a gamma brass structure. The lattice parameter of this gamma brass structure (ao s 9» k 13 A°) differed from Zvickers lattice parameter value (a Q = 9.1*3 A°) by 0.017 A°. This difference in lattice parameter is probably due to variation of lattice parameter with composition, vhich suggests that the compound exists over a small range of composition rather than at a particular composition. Investigation of the magnetic face centred cubic phases in the manganese indium carbon system by Morgan, shoved that the microstructures of this ternary system vere identical vith those illustrated in Goeddel-s masters thesis. This identity does not prove that the ferromagnetic phase found by Goeddel and Yost is the magnetic face centred cubic phase exist-ing In the manganese indium carbon alloys because identical microstructures have been found to exist in a fairly large number of alloys in different alloy systems. Furthermore, some alloys In the manganese nitrogen system 17 (Zvicker ) exhibit the same microstructures as the alloys in Morgans ternary systems and those illustrated in Goeddel1s thesis. However, Goeddel's and Yost's qualitative magnetic results were too erratic to presume that magnetism was due to a single compound in a binary system. In view of the graphite sheaths and other graphite parts used in their apparatus, i t may be presumed that carbon contamination may have taken place. Although Goeddel and Yost took a l l necessary precautions to avoid direct contact between the graphite parts of their apparatus and the alloys, they did not foresee that any remaining oxygen in the furnace after evacuation would oxidize the carbon from the graphite and produce a carburizing atmosphere. Probably the carbon during the melting of the alloys was diffusing interstitially into their alloys. The author feels that i f Goeddel and Yost have concentrated their work on the magnetic phase through X-ray diffraction examination rather than continuing their investigation of the equilibrium diagram by metallographic techniques, they might have been able to explain the erratic qualitative magnetic results and find the true composition of the magnetic phase. (b) Manganese Antimony Carbon Alloys. The melting characteristics, mentioned in the results, of the alloys containing 20 per cent carbon in the manganese antimony carbon system indicate an immiscibility gap exists from approximately 7 P e r cent antimony. The increase in magnetic moment with increasing antimony in the alloys containing 20 per cent carbon suggest that magnetism is due to a manganese antimony phase. This deduction is supported by the identical quantitative magnetic results of the homogenized MngtfSb^&k 8 1 1 4 Mn^b alloys vhich indicate that addition of carbon atoms does not effect magnet-ization. The identical X-ray diffraction patterns of these two alloys in both the c h i l l cast and homogenized states confirm that the magnetic phase is a manganese antimony binary phase, the compound MngSb; furthermore, the identity of these patterns in the homogenized and c h i l l cast states suggests that the same phases exist at high temperatures as ve i l as at low temp-eratures. The reader should be reminded at this point that due to the decomposition of these manganese antimony carbon alloys on quenching from high temperatures, the alloys vere furnace cooled to room temperature after homogenization. As the scattering factor of carbon is very low the role of the carbon atoms in these alloys is impossible to deduce by X-ray diffraction techniques. Visual observations of the Ma^Sh^C^ alloy shoved free carbon particles existing both on the surf ace and in the core of the ingot. The existence of free carbon crust found on slow cooling the MnyoSb^C^Q alloy in the melting furnace. If these visual observations are in error, the carbon would probably exist interstitially in the compound M n23 C6' however, these alloys were found to be strongly magnetic and 18 Wedekind reports the compound Mn^Cg as weakly magnetic. The author, therefore, presumes the existence of ferromagnetism in these alloys is mainly due to the binary manganese antimony compound Mn^ Sb. (c) Manganese Magnesium Alloys. The contradictory X-ray results found when the alloys containing 25 and 33 per cent magnesium vere furnace cooled in the bomb and vhen alloys vith the same composition vere furnace cooled after homogenization suggest contamination in the tube furnaces. The diffraction patterns of the alloys furnace cooled in the bomb shoved alpha manganese in equilibrium 31 with alpha magnesium, while those of the alloys furnace cooled after homogenization shoved alpha manganese in equilibrium with a magnetic face centred cubic phase• It does not seem feasable that slight changes in cooling rate would retain a high temperature phase. The probability of contamination is also supported by the erratic variations in lattice parameter at various temperatures and compositions shown in Table III. Furthermore, the contradictory X-ray results found on quenching the alloys containing 25 and 33 per cent magnesium from 500°C and 600°C after homogen-o izing at 900 C and when alloys with the same composition and quenched from 500°C and 600°C after homogenizing at 500°C and 600°C respectfully, suggest that contamination is occuring at temperatures between 700°C and 900°C. The X-ray patterns of the alloys homogenized at 900°C showed alpha manganese in equilibrium with a magnetic face centred cubic phase while those of the alloys homogenized at 50O°C and 600°C shoved alpha manganese in equilibrium with alpha magnesium. As the tube furnaces did not contain any carbon parts, the author presumed that the contamination was probably not due to carbon but may be due to counter current diffusion of argon with oxygen and nitrogen. The X-ray results of the alloys containing 25 per cent magnesium and heat treated in a nitrogen atmosphere as compared to the X-ray results of the alloy with the same composition and heat treated in vacuo support this hypothesis and suggest that nitrogen is the contaminant. The X-ray pattern of the alloy heat treated in a nitrogen atmosphere showed only a magnetic face centred cubic phase, while those of the alloy heat treated in vacuo shoved beta manganese in equilibrium with alpha magnesium. Although nitrogen may be the contaminant, this assumption does not prove that the magnetic phase is a manganese nitrogen compound. It is possible 32 that a manganese magnesium nitrogen phase may exist. However, the very slight variations in parameter of the alloys ranging from 5 to 25 per cent magnesium heat treated in the powdered form and in a nitrogen atmosphere suggests that the addition of magnesium atoms does not vary the structure and the results support the conclusion that the magnetic phase is the compound Mn^ N. The appearance of magnesium as magnesium oxide in the 25 per cent magnesium alloy and the apparently constant magnetic strength of these alloys provides further support for this conclusion. The largest deviation in lattice parameter of the magnetic face centred cubic phases found in this investigation from the value a Q s 3.855 A° for Mn^ H as reported by Zwicker, is the furnace cooled alloy containing 25 per cent magnesium mentioned in Table III with a lattice parameter value of ^ s 3.869 A°. This large variation in parameter is not understood. The only other face centred cubic lattice with approxim-ately the same lattice parameter is the alloy with the same composition and quenched from 700°C (a Q - 3.867 A°). The films of both alloys showed magnesium oxide lines. The presence of these magnesium oxide lines and the low scattering factors of magnesium and oxygen suggest that a l l the magnesium oxidized, and the magnetic cubic phase is Mni^ N. Visual compar-ison of the intensities of the superlattice lines for these alloys with those for Mn^ N showed no variation, this supports the suggestion that the face centred cubic phase in both alloys is Mn^ K; however, quantitative measurements would be necessary. Another point of interest in the X-ray diffraction examination of the alloy with the large lattice quenched from 700°C and an alloy of the same composition quenched from 600°C is the coexistence of two face centred cubic phases in each alloy. The coexist-ence of two such face centred cubic phase was also found in alloys quenched 33 from the same temperature hut with high magnesium content. In the case of the alloy quenched from 700°C the larger of the two lattices had a parameter value of BQ . 3.867 A° while the smaller of the two lattices had a parameter value of ao > 3.832 A°. The larger lattice which was just discussed was suggested to he Mh^ N. The smaller lattice with a Q r 3.832 A° is too small for a ternary phase containing manganese magnesium and nitrogen. The theoretical value for an ordered face centred cubic lattice of this form or of the form containing only manganese and magnesium atoms is a u 3.84 A°. A random orientation of both the ternary and binary structures may be possible with such a small lattice parameter. The co-existence of two face centred cubic phases in a binary or a ternary system does not seem to be feasible. If the atoms tend to form a face centred cubic structure (or any other structure), a single structure would be more feasible than two identical crystallographic structures with two different atom orientations. The author presumes that the small lattice is probably due to contamination. The coexistence of two face centred cubic lattices in the investigation conducted by Morgan may be due to contamination. The co-existence of the two cubic lattices in the Mhg^  y A l ^ alloy is probably due to nitrogen contamination. The alloy was heat treated in a powdered form at 900°C and furnace cooled. Prom the investigation conducted on the manganese magnesium alloys i t was found that manganese powder nitrides very readily. On examining the films of the magnetic ternary systems, manganous oxide was found in the Mn^IngCgQ and MnQ^Sn^C.^ alloys quenched from 900°C. On the basis of this contamination and the possibility of the existence of magnetic ternary phases containing nitrogen a manganese indium alloy containing 5 per cent indium was heat 3 4 treated at 1100°C for 24 hours in a nitrogen atmosphere. At this temp-erature and composition only gamma manganese exists according to the manganese indium and manganese nitrogen equilibrium diagrams. The alloy was found to be weakly magnetic and the X-ray diffraction film showed gamma manganese and a disordered face centred cubic lattice with a Q = 3.863 A° . This cubic phase is probably manganese nitride; however, the possibility of magnetic ternary phases containing nitrogen should'not be overlooked on the basis of the above results. If Morgan's hypothesis that the ternary magnetic phases are derised from the gamma manganese form is true, then the manganese nitrogen system is the most promising in this respect. The gamma structure in this binary exists over a wide range of composition and temperature as reported by Zwicker. Furthermore manganese nitride (Mn^ N) is face centred cubic and will tend to remain in this form on addition of a third element. The X-ray results of the non contaminated alloys after heat treatment at temperatures below 600°C, where only alpha manganese was found to be in equilibrium with alpha magnesium and the X-ray results of o the alloy heat treated in vacuo at 900 C suggest that only solid solutions of manganese and magnesium are present in this binary system. It may then be concluded that no magnetic phases exist in this system. 35 VI CONCLUSION It is believed that in the formation of the compound Mn^In, the manganese atoms are not far enough apart to produce favourable conditions for magnetism. Magnetism exhibited in manganese antimony carbon alloys is mainly due to the compound Mn2Sb. This suggests that elements in Group Vb which are strongly electronegative to manganese and form very stable binary compounds, do not tend to form ternary compounds containing carbon. Any magnetic phases that may be found with elements more electronegative to antimony in the ternary systems, will: probably be due to a binary compound. In the manganese magnesium system only solid solutions of manganese and magnesium exist. The only probable magnetic phase the manganese solid solution vas found to be non magnetic, which suggests that the addition of magnesium atoms into the solid solution structure did not separate the manganese atoms far enough apart for favourable ferro magnetic conditions. The author suggests that an investigation in the ternary systems containing manganese nitrogen and a third element should be undertaken. Furthermore, a thorough investigation of the manganese carbon system may prove beneficial in understanding the results found in the ternary systems by Morgan. VII APPENDIX (a) Metals Used In this Investigation. MANGANESE INDIUM SYSTEM Metal Type Purity per cent Size of Particles (Tyler Screen Size) mesh. Manganese electrolytic 99-9 • 10 Indium 99.99 • 10 MANGANESE ANTIMONY CARBON SYSTEM Metal Type Purity per cent Size of Particles (Tyler Screen Size) mesh. Manganese electrolytic 99.9 • 10 Antimony 99 * 10 Carbon graphite - 10, • 100 MANGANESE MAGNESIUM SYSTEM Metal Type Purity per cent Size of Particles (Tyler Screen Size) mesh. Manganese electrolytic 99.9 - 65 Magnesium 99.0 - HQ • 65 (b) Vacuum Melting Furnace. To prevent the contamination of the alloys by nitrogen and oxygen, a vacuum melting furnace was used, which is shown diagrammatic-ally in Figure 9* This vacuum melting furnace and its vacuum pump unit 19 were designed and constructed by G. Piercy for his Master's thesis work on the manganese nickel alloys. The rear view of the furnace and the evacuation unit are shown in Figure 10.' The furnace consists of a 1^  inch quartz tube, Ik ins. long, placed coaxially in an induction furnace coil and suspended from a 'T' shaped brass fixture. The bottom opening of the fixture is attached to the quartz tube, while the top opening is scaled off with a watch glass and acts as a window. The third opening of the fixture is used as a changing door and is scaled off by a brass mould during operation. A fourth opening, parallel and in the same plane as the third is attached to the vacuum pumping unit by a 0-ring rotating seal, which allows the furnace to be rotated through an angle of 100 degrees from the vertical for c h i l l casting. The joints between the quartz tube and the rest of the metal parts are sealed by a mixture of octoil and digested rubber in vacuum wax. This mixture has the physical properties of plastecene. A movable rod within and in a coaxial position to the quartz tube allows the charge to be moved vertically in the tube. 'The purpose of the rod is to lower the charge Into the induction coil zone when the crucible with its alundum sheath is placed in the tube, and to raise the molten charge from the melting zone into the 'T* fixture for c h i l l casting. The induction coil had sixteen turns in four inch lengths. The power to the coil was supplied by a 7.5 KW Lepel high frequency unit. 38 Figure 9. - Diagrammatic View of Melting Furnace. ko The evacuation unit consists of a Duo-Seal vacuum pump, diffusion pump, three valses and pressure guages. The valves were a heavy Kinney vacuum valve and two standard water valves. The pressure guages were a rough hydraulic guage, a Pirani guage type PG - IA and a Philips guage type PEG - 1. The diffusion pump and the Philips guage were not used as the forepump was able to maintain a pressure of 20 nicrons of mercury which was satisfactory for this investigation. (c) Tube Furnace and Bomb. Since the very high vapour pressure of magnesium at the melting point of manganese did not permit the use of the vacuum melting unit, a closed bomb technique was employed. The bomb technique was developed by Butters in his investigation of the manganese zinc carbon system. The bomb consisted of a ^  inch seamless steel tube approximately 3 ins. long with mild steel ends welded to the tube. The compact in the bomb is suspended in magnesium oxide powder. The tube furnace consists of a one inch s i l i c a tube placed in a box type heating unit. One end of the tube is sealed off by a rubber stopper with two openings. Through one opening passes a thermocouple encased in £ inch quartz tube and leading to a Wheelco temperature control unit. Through the other opening passes a glass tube connected to a rubber bag. The rubber bag acts as a rough pressure guage. The bag in the blown up state indicates a positive pressure in the tube furnace. The other end of the tube is sealed off by a removable rubber stopper attached to a rubber tube with a clamp. This outlet is connected to a vacuum pump and an argon tank through a 'T* connection. This 'T* connection permits the evacuation and f i l l i n g the quartz tube without a i r contamination. A diagrammatic view of the assembly i s shown i n Figure 11. Figure 11. - Diagrammatic View of Furnace Used i n the Bomb Technique. 42 VIII BIBLIOGRAPHY 1 . Heizeriberg, W., Z. Physik, vol. 49 (1928) p. 619-36 . 2 . Coles, B.R., Thesis, Oxford University, ( l 9 5 l ) . 3 . Pauling, L., Phys. Review, vol. 54, (1938) p. 899 . 4. Slater, J.C., Phys. Review, vol. 36 , (1930) p. 253 . 5 . Bethe H., Handb. d, Physik, vol. 24, (1938) p . 595 -8 . 6 . Morgan, E.R., To be Published. J. Butters, R«, Private Communication. 8 . Goeddel, W.V. and Yost, D.M., Phys. Review, vol. 8 2 , (1951) P« 555* 9 . Zwicker, U., Z. Metallkunde, vol. 41, ( l 9 5 l ) P« 399-400. 1 0 . Barnes, F.A., and Eppelsheimer, D.S., Private Communication (1951) . 1 1 . Schmidt, W., Z. Metallkunde, vol. 1 9 , (1927) P» 452. 1 2 . Grogan, J.D., and Haughton> J.L., Inl. Inst. Metals, vol. 69 , (1943) p.241. 1 3 . Schneider, A., and Scholder, B., Metall, vol. 9 / l 0 , (1950) p . 178 . 14. Siebel, G., Metallwirtschaft, vol. 1 0 , (1931) P» 9 2 3 . 1 5 . Tiner, W., Metals Tech., vol. 12/4, (1945) p. 1 . 1 6 . Bakhmeter, E.F., and Goiovchiner, I.M., Chem. Abstracts, vol. 29 , (1935) p. 7769. 1 7 . Zwicker, U., Z. Metallkunde, vol. 42, ( l 9 5 l ) p. 277 -8 . 1 8 . Wedekind, E., Magnetochemie, (1911) p. 3 3 7 - 9 . 1 9 . Piercy, G., Thesis, University of British Columbia (1952) . 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0106434/manifest

Comment

Related Items