<%BANNER%>

Synthesis and Characterization of BaFeO3 and BiFeO3 Epitaxial Films

Permanent Link: http://ufdc.ufl.edu/UFE0024378/00001

Material Information

Title: Synthesis and Characterization of BaFeO3 and BiFeO3 Epitaxial Films
Physical Description: 1 online resource (122 p.)
Language: english
Creator: Callender, Charlee
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: bafeo3, bifeo3, deposition, epitaxial, ferromagnetism, film, laser, multiferroics, pulsed, solid, solution, strain
Materials Science and Engineering -- Dissertations, Academic -- UF
Genre: Materials Science and Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: SYNTHESIS AND CHARACTERIZATION OF BaFeO3 AND BiFeO3 EPITAXIAL FILMS By Charlee J. Callender Bennett May 2009 Chair: David P. Norton Major: Materials Science and Engineering Much interest exists in perovskite oxide materials and the potential they have in possessing two or more functional properties. In recent years, research on developing new materials with simultaneous ferromagnetic and ferroelectric behavior is the key to addressing possible challenges of new storage information applications. This work examines the fundamental properties of a perovskite oxide, namely BaFeO3, and the investigation of properties of a solid solution between BaFeO3 and BiFeO3. The growth and properties of epitaxial BaFeO3 thin films in the metastable cubic perovskite phase are examined. BaFeO3 films were grown on (012) LaAlO3 and (001) SrTiO3 single crystal substrates by pulsed-laser deposition. X-ray diffraction shows that in situ growth at temperatures between 650-850degreeC yields an oxygen-deficient BaFeO2.5+x pseudo-cubic perovskite phase that is insulating and paramagnetic. Magnetization measurements on the as-deposited BaFeO3 films indicate non-ferromagnetic behavior. Annealing these films in 1 atm oxygen ambient converts the films into a pseudo-cubic BaFeO3-x phase that is ferromagnetic with a Curie temperature of 235 K. The observation of ferromagnetism with increasing oxygen content is consistent with superexchange coupling of Fe+4-O-Fe+4. The effects of anneal conditions on BaFeO3 are studied. X-ray characterization, such as reciprocal space maps, show more complex structure for as-grown BaFeO3-x epitaxial films. Epitaxial films grown at low laser energies are highly crystalline. However, they decompose after annealing. When grown at high laser energies, films exhibit complex structure which cleans up to a single pseudocubic or tetragonal structure upon ex situ anneal in oxygen ambient environment. Superlattices of BaFeO3/SrTiO3 were synthesized to explore the nature of cracking in annealed BaFeO3, which occurs due to large change in lattice parameter. Magnetization of ex situ annealed BaFeO3-x epitaxial films were examined as a function of applied field direction and was not found to have a change in magnetization with direction of field, despite other research claims. Evidence supports that the unusually weak magnetization of BaFeO3-x is attributed to it being structurally and magnetically disordered. Alloys of a solid-solution between BiFeO3 and BaFeO3-x have been successfully created. X-ray characterizations demonstrate alloy epitaxial films via two-target continuous rotation method have been carried all the way to 80% solubility. In addition, alloy films via solid-solution targets method have been successfully fabricated at near both end-member-points and at the half-point showing that the solubility is possible over the entire range of the solid-solution. Bi0.9Ba0.1FeO3 epitaxial films are of high crystalline quality with rocking curves widths of less than 0.22degree, are fully strained, and have highly unusual in-plane and out-of-plane lattice parameters. TEM imaging illustrates that, despite extreme c/a ratios up to 1.26, the films are single phase with sharp interfaces with substrates. SQUID magnetometry was utilized, revealing that the samples are weakly ferromagnetic with a magnetization of 0.2microB per Fe, more than an order of magnitude larger than that of pure BiFeO3. Magnetic hysteresis loops show unfamiliar pinching, signaling a possible breakdown of the helical magnetic ordering in the fully strained samples. BaFeO3-x, though it can be made ferromagnetic, it is a highly complex material. In studying BaFeO3-x s properties, conclusions can be made that its weak magnetization and unusual structure is highly disordered, magnetically and structurally. The creation of a new solid solution (Bi, Ba)FeO3 by two methods shows that a solid solution between BiFeO3 and BaFeO3-x can be synthesized. Specifically the creation of the alloy Bi0.9Ba0.1FeO3-?, shows that one can improve on BiFeO3 s magnetic properties, and more importantly supports the case that BaFeO3-x exhibits magnetic and structural disorder.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Charlee Callender.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Norton, David P.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2009-11-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024378:00001

Permanent Link: http://ufdc.ufl.edu/UFE0024378/00001

Material Information

Title: Synthesis and Characterization of BaFeO3 and BiFeO3 Epitaxial Films
Physical Description: 1 online resource (122 p.)
Language: english
Creator: Callender, Charlee
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: bafeo3, bifeo3, deposition, epitaxial, ferromagnetism, film, laser, multiferroics, pulsed, solid, solution, strain
Materials Science and Engineering -- Dissertations, Academic -- UF
Genre: Materials Science and Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: SYNTHESIS AND CHARACTERIZATION OF BaFeO3 AND BiFeO3 EPITAXIAL FILMS By Charlee J. Callender Bennett May 2009 Chair: David P. Norton Major: Materials Science and Engineering Much interest exists in perovskite oxide materials and the potential they have in possessing two or more functional properties. In recent years, research on developing new materials with simultaneous ferromagnetic and ferroelectric behavior is the key to addressing possible challenges of new storage information applications. This work examines the fundamental properties of a perovskite oxide, namely BaFeO3, and the investigation of properties of a solid solution between BaFeO3 and BiFeO3. The growth and properties of epitaxial BaFeO3 thin films in the metastable cubic perovskite phase are examined. BaFeO3 films were grown on (012) LaAlO3 and (001) SrTiO3 single crystal substrates by pulsed-laser deposition. X-ray diffraction shows that in situ growth at temperatures between 650-850degreeC yields an oxygen-deficient BaFeO2.5+x pseudo-cubic perovskite phase that is insulating and paramagnetic. Magnetization measurements on the as-deposited BaFeO3 films indicate non-ferromagnetic behavior. Annealing these films in 1 atm oxygen ambient converts the films into a pseudo-cubic BaFeO3-x phase that is ferromagnetic with a Curie temperature of 235 K. The observation of ferromagnetism with increasing oxygen content is consistent with superexchange coupling of Fe+4-O-Fe+4. The effects of anneal conditions on BaFeO3 are studied. X-ray characterization, such as reciprocal space maps, show more complex structure for as-grown BaFeO3-x epitaxial films. Epitaxial films grown at low laser energies are highly crystalline. However, they decompose after annealing. When grown at high laser energies, films exhibit complex structure which cleans up to a single pseudocubic or tetragonal structure upon ex situ anneal in oxygen ambient environment. Superlattices of BaFeO3/SrTiO3 were synthesized to explore the nature of cracking in annealed BaFeO3, which occurs due to large change in lattice parameter. Magnetization of ex situ annealed BaFeO3-x epitaxial films were examined as a function of applied field direction and was not found to have a change in magnetization with direction of field, despite other research claims. Evidence supports that the unusually weak magnetization of BaFeO3-x is attributed to it being structurally and magnetically disordered. Alloys of a solid-solution between BiFeO3 and BaFeO3-x have been successfully created. X-ray characterizations demonstrate alloy epitaxial films via two-target continuous rotation method have been carried all the way to 80% solubility. In addition, alloy films via solid-solution targets method have been successfully fabricated at near both end-member-points and at the half-point showing that the solubility is possible over the entire range of the solid-solution. Bi0.9Ba0.1FeO3 epitaxial films are of high crystalline quality with rocking curves widths of less than 0.22degree, are fully strained, and have highly unusual in-plane and out-of-plane lattice parameters. TEM imaging illustrates that, despite extreme c/a ratios up to 1.26, the films are single phase with sharp interfaces with substrates. SQUID magnetometry was utilized, revealing that the samples are weakly ferromagnetic with a magnetization of 0.2microB per Fe, more than an order of magnitude larger than that of pure BiFeO3. Magnetic hysteresis loops show unfamiliar pinching, signaling a possible breakdown of the helical magnetic ordering in the fully strained samples. BaFeO3-x, though it can be made ferromagnetic, it is a highly complex material. In studying BaFeO3-x s properties, conclusions can be made that its weak magnetization and unusual structure is highly disordered, magnetically and structurally. The creation of a new solid solution (Bi, Ba)FeO3 by two methods shows that a solid solution between BiFeO3 and BaFeO3-x can be synthesized. Specifically the creation of the alloy Bi0.9Ba0.1FeO3-?, shows that one can improve on BiFeO3 s magnetic properties, and more importantly supports the case that BaFeO3-x exhibits magnetic and structural disorder.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Charlee Callender.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Norton, David P.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2009-11-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024378:00001


This item has the following downloads:


Full Text

PAGE 1

1 SYNTHESIS AND CHARACTERIZATION OF BaFeO3 AND BiFeO3 EPITAXIAL FILMS By CHARLEE J. CALLENDER BENNETT A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2009

PAGE 2

2 2009 Charlee J. Callender Bennett

PAGE 3

3 To Jesus, because He is the One

PAGE 4

4 ACKNOWLEDGMENTS First and forem ost, to my Lord and savior Je sus Christ, who is God on Earth. I thank Him for all that has come to pass in my life, for I know all of it is in vain if it is not for Him. I want to thank and bless my advisor and mentor, Asso ciate Dean David P. Norton. God has moved through him to help me become the scholarly pers on I am today; his true belief in my abilities always surpassed my own. I want to praise my committee, Dr. Stephen Pearton, Associate Dean Cammy Abernathy, Dr. Simon Phillpot and Dr. Arthur Hebard. I th ank them for valuable talks, amiable conversations, and supportive wisdom duri ng my tenure at the University of Florida. I give special thanks to Dr. Hans Christe n, my mentor at Oak Ri dge National Laboratory, for granting me an extraordinary opportunity to conduct research with wo rld-known scientists. I have absolutely no regrets of my time at ORNL and truly appreciate the relationships built with the Thin Films and Nanostructures research group and several others. I thank Dr. John Budai for valuable perspectives on x-ray diffraction analysis. I want to take an exclusive moment to praise my family: my parents, Mr. Joseph Callender and Mrs. Sybil Callender, without who I would have ceased to be. Their words: Get the highest degree you can has always played in the forefront of my mind, even when I didnt know the first thing about higher education when en tering college. They ar e truly proud to have produced the first Dr. Callender (Bennett) in the hist ory of the Callender ex tended family. My sisters and brother, Wayne, Shana, and Camille, have been a happy challenge as I embarked on this journey of life through this point, and I was pleased to have graduated with Camille (youngest), who achieved her Masters degr ee when I achieved the Doctorate. My husband, Mr. Andrew Bennett has sacrificed the totality of himself to support me to this point, and I forever praise him and am excited about the educational jour ney that awaits him. I thank God again for joining Andrew and me as one, He knew the path for us, and I pray He

PAGE 5

5 continues to order our steps in His Word. Th ank you to my extended family and water-relatives in the United States, Jamaica, England, and Ca nada. I whole-heartedly agree that it takes a village to properly raise a child into what God has for their live s. Each of their encouraging words and love has helped shape my life, and my appreciation for valuable relationships. I want to officially acknowledge my present and past research group members for valuable group interactions, friendship and support: Dr. Hyung-Jin Bae, Dr. Yuanjie Li, Dr. Seemant Rawal, Dr. Jean-Marie Erie, Dr. Mat Ivill, Dr Li-Chia Tien, Dr. HyunSik Kim, Dr. Lil-Cherng Leu, Ryan Pate, Patrick Sadik, Fernando Lugo, Zi vin Park, and Joe Cianfrone. My sincere gratefulness to Dr. Bae and Dr. Iv ill, both whom were instrumental to me developing valuable research skills early on in the Norton Research Labs. I want to take a moment to reflect on our 44th President, Barack H. Obama, First Lady Michelle Obama, and their First family. Thr oughout my dissertation process, his advancement through Senate, presidential primarie s, candidacy and U. S. Presiden t has given me no excuse to not complete what I have started. He and his fa milys trials and successes motivate me to take bolder steps beyond the PhD. I also want to give words of love to two of my mentors, Dr. Amy Lovell and Dr. Emilia Hodge, who consistently ra ise my personal value when it has been pulled by others. Dr. Danyell Wilsonall I can say is I will always thank Jesus for putting her in my life, my sister in Christ!! Abiding Faith CC is the church home I will never forget, bless them! Finally, thanks to the MSE Dept, especially Academic Advising: Ms. Doris, Ms. Jennifer, and Ms. Martha. They perceived me first as a fi t for a top ten graduate program, and helped support me throughout the years. My apprec iation goes to NSF-AGEP, BOE, OGMP, and NACME for emotional and financial support. Finally, many thanks goes to ORNL and MAIC for use of their facilities. To all (especially those not listed due to space limits) I am so grateful.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ............................................................................................................... 4LIST OF TABLES ...........................................................................................................................9LIST OF FIGURES .......................................................................................................................10ABSTRACT ...................................................................................................................... .............13 CHAP TER 1 INTRODUCTION .................................................................................................................. 162 REVIEW OF MULTIFERROIC MATERIALS ....................................................................18Introduction .................................................................................................................. ...........18Magnetism ..............................................................................................................................18Diamagnetism .................................................................................................................. 18Paramagnetism ................................................................................................................ 19Ferromagnetism ............................................................................................................... 20Antiferromagnetism ......................................................................................................... 20Ferrimagnetism ................................................................................................................ 21Ferroelectrics ................................................................................................................ ..........22Ferroelectricity ................................................................................................................22Antiferroelectricity .......................................................................................................... 23BaFeO3 ....................................................................................................................................23BiFeO3 ....................................................................................................................................24Multiferroics ................................................................................................................. ..........25Introduction .................................................................................................................. ...25Single Phase Multiferroics ..............................................................................................26Multiferroic Composites .................................................................................................. 27Multiferroic Materials via Artificial Superlattices .......................................................... 27Multiferroic Solid-Solut ion Epitaxial Films .................................................................... 293 METHODOLOGY ................................................................................................................. 36Synthesis of Thin Films ....................................................................................................... ...36Pulsed Laser Deposition ..................................................................................................36Introduction .............................................................................................................. 36Why pulsed laser deposition? ...................................................................................36Components .............................................................................................................. 37Laser ......................................................................................................................... 37Laser-target interaction ............................................................................................ 38Optics ....................................................................................................................... 38Deposition chamber .................................................................................................. 38

PAGE 7

7 Targets ...................................................................................................................... 39Growth procedure ..................................................................................................... 40Two -target continuous rotation method .................................................................. 40Characterization .............................................................................................................. ........41Superconducting Quantum Interferen ce Device Magnetometry (SQUID) ..................... 41Introduction .............................................................................................................. 41Theory ...................................................................................................................... 41Magnetic Property Measurement System (MPMS) components ............................. 42Sample preparation ................................................................................................... 43Sample mount ........................................................................................................... 43Sample measurement ................................................................................................ 44X-Ray Diffraction (XRD) ................................................................................................ 46Introduction .............................................................................................................. 46Braggs Law (X-ray Theory) ....................................................................................47Reciprocal space maps ............................................................................................. 48Other scans ...............................................................................................................48Components used in the study .................................................................................. 49Scanning Electron Microscopy (SEM) ............................................................................ 49Scanning electron microscopy operation ................................................................. 49Sample preparation and mounting ............................................................................ 50X-Ray spectral measurement via electr on dispersive spectrometer (EDS) ..............504 FERROMAGNETISM IN PSEUDO-CUBIC BaFeO3 EPITAXIAL FILMS ........................ 53Introduction .................................................................................................................. ...........53Experimentation ............................................................................................................... .......54Results and Discussion ........................................................................................................ ...55As-Deposited BaFeO3-x Epitaxial Films ..........................................................................55Ex situ anneal treatment of BaFeO3-x Epitaxial Films ..................................................... 56Magnetic Properties of Annealed BaFeO3-x Epitaxial Films ........................................... 56Conclusion .................................................................................................................... ..........585 STRUCTURAL CHALLENGES WITH BaFeO3-X EPITAXY .............................................64Introduction .................................................................................................................. ...........64Experimentation ............................................................................................................... .......64Results and Discussion ........................................................................................................ ...65BaFeO3 Epitaxial Films Grown at Low Laser Energy .................................................... 65BaFeO3 Epitaxial Films Grown at High Laser Energy ................................................... 66Effects of film thickness on magnetic properties of BaFeO3 ...................................67Effects of oxygen and nitrous oxide pa rtial pressure on structure of BaFeO3 .........68Magnetization in Different Directions .............................................................................69Artificial Superlattices of SrTiO3/BaFeO3 ......................................................................70Conclusion .................................................................................................................... ..........71

PAGE 8

8 6 SYNTHESIS AND PROPERTIES OF (Bi,Ba)FeO3 SOLID SOLUTION THIN FILMS .... 88Introduction .................................................................................................................. ...........88Experimentation ............................................................................................................... .......89Synthesis of (BixBa1-x)FeO3-x Films via Continuous Alte rnating Target Rotation ......... 89Synthesis of (BixBa1-x)FeO3-x Films via Solid Solution Targets ..................................... 89Results and Discussion ........................................................................................................ ...90(BixBa1-x)FeO3-x Films via Continuous Alte rnating Target Rotation ..............................90(BixBa1-x)FeO3 Films via Solid Solution Targets ............................................................ 92Conclusion .................................................................................................................... ..........977 CONCLUSION .................................................................................................................... .112LIST OF REFERENCES .............................................................................................................115BIOGRAPHICAL SKETCH .......................................................................................................122

PAGE 9

9 LIST OF TABLES Table page 2-1 Single phase multiferroics with vario us properties. ........................................................... 343-1 Excimer Laser Operating Wavelengths ............................................................................. 525-1 Comparison of lattice constant mism atch with the two phases of BaFeO3-x .....................826-1 Measured c-axis d-spacings of substrate and corresponding films, taken from twotheta scan. ................................................................................................................... ......1046-2 Measured and calculated in -plane and out of plane lat tice parameters, and unit cell volumes for selected epitaxial films ............................................................................... 108

PAGE 10

10 LIST OF FIGURES Figure page 2-1 Schematic of a perovskite structur e, below and above the ferroelectric TC. ..................... 332-2 Schematic of self-assembled nanostructu re thin film aligned perpendicular to substrate. ............................................................................................................................352-3 Variation of layering in an artificial superlattice.. ............................................................. 353-1 Schematic of Laser MBE Growth Chamber Setup ............................................................ 524-1 X-ray Diffraction data of as-deposited BaFeO3-x epitaxial films with increasing temperature ................................................................................................................... .....594-2 Magnetization versus applied magnetic fi eld for as-deposited BaFeO3-x grown at 750C and 10 mTorr .......................................................................................................... 604-3 X-ray diffraction scans for BaFeO3-x films on LaAlO3, both as deposited and after annealing in 1 atm oxygen for 1 hr at 900C ..................................................................... 614-4 Magnetization versus applied magnetic field for BaFeO3 film grown on SrTiO3 at 600C and 100 mTorr plus annealing in 1 atm oxygen for 1 hr at various temperatures .................................................................................................................. .....624-5 Field-cooled and zero fiel d-cooled magnetization as function of temperature for the BaFeO3-x film after annealing in 1 atm oxygen for 1 hr at 900C .....................................624-6 Magnetization results for BaFeO3 films on SrTiO3.. .........................................................635-1 Two-Theta-Omega scans of as-grown BaFeO3-x thin films at various pressures. ............. 735-2 Omega Rocking Curves show FWHM of (002) peaks of the films at various pressures. .................................................................................................................... ........745-3 Optical Light Microscope images of su rfaces of as-deposited and annealed BaFeO3 films. ........................................................................................................................ ..........755-4 Omega Rocking Curves of as-deposited BaFeO3 epitaxial films at different thicknesses. ........................................................................................................................765-6 Plot of remnant magnetization, Mr versus film thickness (in nm). .................................... 785-7 Reciprocal Space Maps ..................................................................................................... .795-8 As-deposited BaFeO3-x at 100mTorr PN2O .........................................................................80

PAGE 11

11 5-9 Magnetization versus Magnetic field data of BaFeO3-x measured parallel and perpendicular to the crystal axis ........................................................................................ 835-10 X-ray diffraction of [SrTiO3/BaFeO3] artificial superlattices on LAO and STO substrates. Inset is reciprocal space map of superlattice on LAO. .................................... 845-11 X-ray diffraction data of as-grown and a nnealed [STO/BFO] artifi cial superlattices. Attempted periodicity is 4nm. ..................................................................................... 855-12 Schematic of substrate holder after film deposition with substrates attached. Substrates are labeled with corresponding periodicities ( ) of the [STO/BFO] superlattice growth. .......................................................................................................... ..865-13 Optical microscope images of artificial superlattices. ....................................................... 876-1 Structural data of three BixBa1-xFeO3 ceramic targets. ...................................................... 986-2 X-ray diffraction data fo r varied thicknesses of Bi0.5Ba0.5FeO3-x on DyScO3 substrates. ................................................................................................................... ........996-3 X-ray diffraction (2-theta scan) for Bi0.5Ba0.5FeO3-x alloy films at various thicknesses with SmScO3 buffer. ........................................................................................................1006-4 X-ray diffraction data for Bi0.5Ba0.5FeO3-x alloy films at various thicknesses. Omega rocking curves of thin films show ing strain/relaxation behavior. ....................................1006-5 X-ray diffraction data for solid solution (BixBa1-x)FeO3 grown by pulsed-laser deposition. ................................................................................................................... .....1016-6 Reciprocal space maps for selected (Bix Ba1-x)FeO3 films on LaAlO3 with SmScO3 buffer. ....................................................................................................................... ........1026-7 In-plane and out-of-plane lattice parame ters extracted from the reciprocal space maps. ......................................................................................................................... .......1036-8 C/A ratios calculated from lattice cons tants extracted from the reciprocal space maps. ......................................................................................................................... .......1036-9 Bi0.9Ba0.1FeO3-x epitaxial films grown on f our different substrates. ................................ 1046-10 Omega rocking curves with measured widt hs illustrate highly crystall ine phases. ......... 1056-11 Reciprocal space maps of Bi0.9Ba0.1FeO3-x epitaxial films grown on four different substrates .................................................................................................................... ......1066-12 C-axis versus a-axis lattice parameters of Bi0.9Ba0.1FeO3 strained films. ........................ 1096-13 C/A ratio of Bi0.9Ba0.1FeO3 strained films. ...................................................................... 109

PAGE 12

12 6-14 Z-contrast STEM image of Bi0.9Ba0.1FeO3 (light area) on LSAT (dark area). ................ 1106-15 Z-contrast STEM image of Bi0.9Ba0.1FeO3 (light area) on LAO (dark area). .................. 1106-16 Magnetization vs Field data of Bi0.9Ba0.1FeO3-x epitaxial films grown on three out of four different substrates.. .................................................................................................111

PAGE 13

13 Abstract Of Dissertation Pres ented To The Graduate School Of The University Of Florida In Partial Fulfillment Of The Requirements For The Degree Of Doctor Of Philosophy SYNTHESIS AND CHARACTERIZATION OF BaFeO3 AND BiFeO3 EPITAXIAL FILMS By Charlee J. Callender Bennett May 2009 Chair: David P. Norton Major: Materials Scie nce and Engineering Much interest exists in perovskite oxide materials and the potential they have in possessing two or more functional properties. In recent y ears, research on developing new materials with simultaneous ferromagnetic and ferroelectric behavior is the key to addressing possible challenges of new storage information applica tions. This work examines the fundamental properties of a perovskite oxide, namely BaFeO3, and the investigation of properties of a solid solution between BaFeO3 and BiFeO3. The growth and properties of epitaxial BaFeO3 thin films in the metastable cubic perovskite phase are examined. BaFeO3 films were grown on (012) LaAlO3 and (001) SrTiO3 single crystal substrates by pulsedlaser deposition. X-ray diffracti on shows that in situ growth at temperatures between 650-850C yi elds an oxygen-deficient BaFeO2.5+x pseudo-cubic perovskite phase that is insulating and para magnetic. Magnetization measurements on the asdeposited BaFeO3 films indicate non-ferro magnetic behavior. Annealing these films in 1 atm oxygen ambient converts the film s into a pseudo-cubic BaFeO3-x phase that is ferromagnetic with a Curie temperature of 235 K. The observa tion of ferromagnetism with increasing oxygen content is consistent with superexchange coupling of Fe+4-O-Fe+4.

PAGE 14

14 The effects of anneal conditions on BaFeO3 are studied. X-ray ch aracterization, such as reciprocal space maps, show more co mplex structure for as-grown BaFeO3-x epitaxial films. Epitaxial films grown at low laser energies are highly crystalli ne. However, they decompose after annealing. When grown at high laser energies, films e xhibit complex structure which cleans up to a single pseudocubi c or tetragonal structure upon ex situ anneal in oxygen ambient environment. Superlattices of BaFeO3/SrTiO3 were synthesized to explore the nature of cracking in annealed BaFeO3, which occurs due to large change in lattice parameter. Magnetization of ex situ annealed BaFeO3-x epitaxial films were ex amined as a function of applied field direction and was not found to have a change in magnetization with direction of field, despite other research claims. Evidence su pports that the unusually weak magnetization of BaFeO3-x is attributed to it being struct urally and magnetically disordered. Alloys of a solid-solution between BiFeO3 and BaFeO3-x have been successfully created. X-ray characterizations demonstrate alloy epita xial films via two-targ et continuous rotation method have been carried all the way to 80% so lubility. In addition, alloy films via solidsolution targets method have been successfully fabricated at near both end-member-points and at the half-point showing that the so lubility is possible ove r the entire range of the solid-solution. Bi0.9Ba0.1FeO3 epitaxial films are of high crystalline qua lity with rocking curves widths of less than 0.22, are fully strained, and have highl y unusual in-plane and out-of-plane lattice parameters. TEM imaging illustrates that, despite extreme c/a ratios up to 1.26, the films are single phase with sharp interfaces with substrat es. SQUID magnetometry was utilized, revealing that the samples are weakly ferro magnetic with a magnetization of 0.2 B per Fe, more than an order of magnitude larger than that of pure BiFeO3. Magnetic hysteresis loops show unfamiliar

PAGE 15

15 pinching, signaling a possible br eakdown of the helical magnetic ordering in the fully strained samples. BaFeO3-x, though it can be made ferromagnetic, it is a highly complex material. In studying BaFeO3-xs properties, conclusions can be made that its weak magnetization and unusual structure is highly disord ered, magnetically and structurall y. The creation of a new solid solution (Bi, Ba)FeO3 by two methods shows that a solid solution between BiFeO3 and BaFeO3-x can be synthesized. Specifically the creation of the alloy Bi0.9Ba0.1FeO3, shows that one can improve on BiFeO3s magnetic properties, and more impor tantly supports the case that BaFeO3-x exhibits magnetic and structural disorder.

PAGE 16

16 CHAPTER 1 INTRODUCTION Multif erroic materials have been of interest due to their possible use in improvement of new kinds of storage devices. These materials have two or more ferroic pr operties that coexist in the same phase. The properties may or may not be coupled. Multiferroic materials are an example of a broader class of multifunctional or smart materials which combine several properties in the same material and in turn th is produces a new phenome non that is more than just the total sum of its parts.1 The most interesting multiferroic materials are those that can exhibit ferroelectricity and ferromagnetism in the same phase. The applications for these materials are numerous, one example being an electric field switchable magnetization. Multiferroics were discovered many decades ago, but the resurgence in interest of these materials comes mainly from three factors that relate to each other, as mentioned by W. Eerenstein, et al .2 With regards to the first factor, Nicola Spaldin wrote a review article1, 3 discussing the requirements for ferroelectricity and ferromagnetism to exist simultaneously in oxides and hence theorized why there are so few magnetic ferroelectrics (or multiferroics).2, 3 This declaration posed a challenge to scientists to address this problem with many investigations of new materials. Secondly, in the past, techniques able to investigate on an atomic scale were not available at the time the earliest multiferroic s were discovered to probe deep into their complex properties, but as of recent years many characterization and analysis techniques are readily available for the study of new materials to hone their propert ies. Last, Eerenstein cites that there is a relentless drive in the direction of bette r technology that is fueled by investigation of new materials, to be eventual ly utilized as improved transdu cers, magnetic field sensors, and most importantly and presently, information storage technology.2, 4, 5

PAGE 17

17 Before beginning to apply these smart material technologies, there is great need to address fundamental properties of current materials, and to use that knowledge to improve them by way of, for example, making new materials in place of th e existing structures in such a way that the resulting properties are complementary. This work examines the fundamental properties of certain perovskite oxides and the effect on their properties when jo ined together. Chapter 2 gives an overview of magnetism, ferroelectrics, and multiferroics. Then a brief history of perovskite materials of study, BaFeO3 and BiFeO3, is elucidated upon with the intent of studying the mixing of the two. In addition, the ch apter introduces a discussion of the history of multiferroic materials, and past and current study of different single phase, composite, heterostructure, and solid solution materials. Chapter 3 introduces the chosen form of sample synthesis: Pulsed Laser Deposition, its advantages and components, and the proce ss of film deposition. In addition, various characterization methods for sample analysis ar e briefly discussed. I nvestigation of BaFeO3 and its structural and magnetic properties, which differ from that of the bulk, is examined in Chapter 4. The discovery of its magnetic properties upon oxygenation is discussed. Chapter 5 revisits BaFeO3 to look more intimately at complexities of its structural properties via x-ray diffraction, and the effects of growing it in di fferent conditions to make it diffi cult to be easily convertible to a magnetic material. Chapter 6 introduces th e mixing (and discovery) of the synthesis and properties of a single phase solid solution between BaFeO3 and BiFeO3. The structural and magnetic properties of a specific alloy composition are discussed as a function of synthesis and substrate choice. Chapter 7 ties all of the result s and analysis of each e xperiment together to complete the bulk of this work. It further addr esses future challenges that the next research scientist may embark on, if he or she so choose.

PAGE 18

18 CHAPTER 2 REVIEW OF MULTIFERROIC MATERIALS Introduction As m entioned in the previous chapter, multife rroics are interesting materials. In this chapter, the properties that come together to ma ke a multiferroic material are briefly defined, and then history and reviews of multiferroic materials are discusse d. In addition, two materials BaFeO3 and BiFeO3are introduced. Magnetism Am pre postulated more than a century ago that charge currents are responsible for magnetism in solids.6, 7 A few fundamental concepts and properties of magnetism will be discussed, including brief defin itions of diamagnetism, paramagnetism, ferromagnetism, and antiferromagnetism. In addition, some applicat ions of the discussed magnetic properties are provided. Diamagnetism There is a c hange in orbital motion that is du e to an externally applied magnetic field that occurs in materials, including those whose electr on shells are completely filled. This is the diamagnetic effect.6 All materials possess the diamagnetic effect, though it is weak because it is over-shadowed by much stronger magnetic interac tions that are briefly discussed below. Materials and/or atoms that are diamagnetic ha ve a small and negative magnetic susceptibility, m. Materials with closed electr on (sub)shells exhibit diamagnetism.8 Thus, diamagnetic materials do not have a permanent magnetic moment and by themselves are not useful in a wide range of magnetic applications. Nonetheless, there are applications where alloys of diamagnetic and paramagnetic (defined in the next section) materials have a distinct composition at which the magnetism cancels out and the susceptibility, is zero. At this composition, the alloy is not

PAGE 19

19 affected by external magnetic fields, and it can be used in devices designed to make delicate magnetic measurements.6 Paramagnetism Materials that have net magnetic moments, but are only weakly coupled, are known as paramagnets.8 Transition metal salts exhibit paramagnetism from a transition metal cation having a partially filled d shell, and the anions make certain there is enough spatial separation between cations, making the magnetic intera ctions between neighboring cations weak.6 Dilute gases, rare earth salts and elements have an addi tional source of paramagnetism, which is due to the magnetic moment of the orbiting electrons. Without a field, there is a net magnetization of zero, but when a field is applied, the magnetic moments align somewhat in the general field direction. Thermal changes counteract their a lignment, making electron-orbit paramagnetism temperature dependent.7 Some paramagnetic materials can obe y the Curie law shown in Eq. 2-1: T C ,(2-1) which states that the susceptibil ity is inversely proportional to the absolute temperature T, where C is the Curie constant. Other materials obs erve the Curie-Weiss law shown by Eq. 2-2: T C,(2-2) where is a constant that has the same units at C.8,9 In Eq. 2-2, when T = there is a divergence in the susceptibility that corresponds to a phase transition to a spontaneously ordered magnetic phase.6 Like diamagnetic materials, only a few applications exploit paramagnetism due to the lack of net permanent magnetic mome nt. However, paramagnets allow scientists to study the electronic properties of substances with atomic magnetic moments without the hindrance of strong coupling effects.

PAGE 20

20 Ferromagnetism In the earlier section where param agnetism was briefly discussed, Eq. 2-2 showed the Curie-Weiss behavior for several paramagnetic materials. When is positive, then the material is below the Curie Temperature, TC, a critical temperature below which a paramagnetic material exhibits ferromagnetism. Origins of ferro magnetism come from the quantum mechanical exchange interaction between component atoms resulting in areas of the material possessing permanent magnetization.6-8 Ferromagnetic materials have strong permanent magnetizations even in the absence of an externally app lied magnetic field. The magnetic susceptibility m is positive and depends on the history of the applied field. A largely nonlinear relationshi p exists between the externally applied magnetic field oH and the magnetization M and at very high magnetic fields (depending on the material) the magnetization M saturates, called the saturation magnetization Ms.7 When the magnetic field reduces to zero, the material has a remnant magnetization Mr which is employed in many applications of permanent magnets. Further reducing the applied field below zero causes the magnetization M to eventually reach zero, reaching the coercive field Hc at M = 0.8 Antiferromagnetism Antiferrom agnets are made of two identical interpenetrating magnetic sublattices with antiparallel moments. This gives a net zero magnetic moment, with a susceptibility m that is small and positive and a linear magnetization in an applied field.6 This is similar to the response of paramagnetic materials. The susceptibility of Antiferromagnets is direction dependent. Antiferromagnets have a phase transition te mperature called the Nel Temperature TN. Above this temperature the material behaves like a paramagnet. Antiferromagnetism is difficult to distinguis h, even with the use of a Superconducting Quantum Interference Device (SQUID) magnetometr y (discussed in Chapter 3). However the

PAGE 21

21 magnetic structure of antiferromagnets can be determined and quantified using neutron diffraction experimental methods. Most origins of antiferroma gnetic ordering come from the similar quantum mechanical exchange forces of that in ferromagnets except that antiferromagnets possess a negative molecular field, causing the spins to align anti-parallel.6 Superexchange A common way to predict antiferromagnetic orde ring is to consider a system, such as Fe3O4, where there are linear chains of Fe3+-O2--Fe3+ linkages. The oxygen outer electron shell is partially filled with 6 electrons and so hybridization occurs by the donation of electrons from O2ion onto the all vacant sp in up (or all spin down) orbitals of the Fe3+ ions. This occurs throughout the chains leaving a zero net magnetic moment from this oxygen mediated interaction.10 Ferrimagnetism Ferrim agnetic materials are similar to ferroma gnetic materials in that they exhibit spontaneous magnetization below the critical temperature, TC called the Curie temperature, even in the absence of an externally applied field. Ferrimagnetic materials are also similar to antiferromagnetic materials in that they also have exchange coupling between neighboring magnetic ions leading to moments arranging in antiparallel alignment. However, the ferromagnets still have a net permanent magnetic moment, and still have large positive susceptibility. The two magnetic sublattices diffe r in that one sublatti ce will have a greater magnetic moment than that of the othe r, yielding an overall magnetic moment.6-8 Ferrimagnets have many unique applications, due to the fact that most of them are insulating (in contrast, most ferromagnets are metals).

PAGE 22

22 Ferroelectrics Of the 32 crystal classes, there are 11 centrosymmetric (non polar), non-piezoelectric groups. This leaves 21 non-centrosymmetric gr oups, which are piezoelectric, and in these 21 there are 10 polar piezoelectrics. A piezoelectric material is one that undergoes a change in polarization (as well as becoming electrical ly polarized) when it becomes strained.11 In opposition, after applying an electr ic field, the crystal will be st retched or compressed depending on the electric fields orientation to the polarizat ion of the crystal. Crystals in the 10 polar groups are considered pyroelectric in that they have a property that is inherent in their structurespontaneous polarization. In addition, their polarizati on changes with temperature. When in a polar state these crystals possess a permanent dipole.12 Ferroelectricity Of the pyroelectric materials, some have an additional property where, if an electric field is applied, the direction of polarization can change.11 This property is te rmed ferroelectricity. Often a change in the polarization orientation of ferroelectrics results in a change in shape. The most widely studied ferroelectrics have the perovskite structure. The noncentrosymmetric structure in perovskites is reach ed by A or B cations (or both) offcenter relative to the oxygen anions, shown in Figure 2-1, and the spontaneous polarization comes from the electric dipole moment created by this offcentering. In applications, for almost twenty years ferro electrics have been a leading candidate for nonvolatile memories, but one of th e hindrances from widespread ap plication of these materials, is the fatigue that arises from multiple cycling of th ese materials. This limits the reliability of the devices, though oxide electrodes ha ve helped ease this fatigue.13 There is great interest in nondestructive read-out devices, ferroelectric fieldeffect transistors (FFETs), and ferroelectricferromagnetic structure devices for memory applications.

PAGE 23

23 Antiferroelectricity In som e of the perovskite materials, the st ructural distortions th at occur at the Curie temperature, TC, are not ferroelectric, but are instead an tiferroelectric. An antiferroelectric materials ions will displace in ways that ar e analogous to antiferromagnetic materials, but instead there is the creation of two alternating dipole sublattices that ha ve equal polarization, but in opposite directions.12 Thus, the net polarization is zero, a nd the dielectric constant changes at TC. BaFeO3 BaFeO3 is an ABO3 type perovskite material. In bul k, it exists as a primarily hexagonal perovskite structure with lattice parameter a = 0.568 nm and c = 1.386 nm.14-17 In some cases reports have shown it to have an antiferromagne tic (low temperature state) to ferromagnetic transition around 160 K as well as another transition from ferro magnetic to the paramagnetic state around 250 K.18-20 The mechanism by which BaFeO3 shows ferromagnetism is via superexchange interaction energy of the Fe3+-O2-Fe3+ linkage.16 When grown as thin films, BaFeO3 can take on a pseudo-cubic perovskite structure (a = 0.412 nm) as reported by Matsui, et al 21, 22 with iron in the center of oxygen octahedra linking to neighboring octahedra via sharing corners. Depending on partial pr essure of oxygen and temperature conditions, it can be oxygen deficient, meaning the stoichiometric formula deviates from BaFeO3 and gets closer to BaFeO2.5. The impact of oxygen defici encies have been reported to influence its magnetic properties, where with in crease in oxygen deficiency there is a decrease in ferromagnetism.18 The mechanism by which this increase occurs is speculated to be a change in the valence state of Fe ions ; an increase in oxygen deficien cy leads to a decrease in Fe4+ and an increase in Fe3+ ions. Evidence suggests that this expl anation is sound due to the associated increase in the materials lattice spacing due to decrease in oxygen vacancies (more metal ions

PAGE 24

24 with higher valence stat e have larger radius).17-20 BaFeO3s complexities, when grown as thin films are discussed in Chapter 5 to yield unders tanding of the nature of its structural and magnetic properties. BaFeO3 is an interesting material to investigate because it is ferromagnetic and it can possibly be combined with BiFeO3 to yield even more intere sting properties in the area of multiferroism. BiFeO3 BiFeO3 is a multiferroic material that exhibits ferroelectricity and antiferromagnetism in the same phase. It is currently one of the most studied multiferroic materials. Bulk BiFeO3 exists as a distorted perovskite structur e, with rhombohedral polar space group R3c.23, 24 The space group R3c allows for BiFeO3 to exhibit ferroelectricity.25-27 The rhombohedral unit cell contains two connected BiFeO3 perovskite formula units, with the two oxygen octahedra tilted by + and 13.8.28 As of recently, in bul k, single crystal BiFeO3 can exhibit a spontaneous polarization in excess of 100 C cm-2.29 The ferroelectricity in BiFeO3 can be explained by lone pair mechanisms of the Bi ions.30 Its polarization occurs alo ng the [111] direction, and it is ferroelectric below Curie temperature, TC = 1143 K.31, 32 As mentioned before, BiFeO3 is also magnetic, exhibiti ng G-type antiferromagnetism below a Neel Temperature of 643 K, also having weak ferromagnetism present.31, 33, 34 It was found that BiFeO3s weak ferromagnetism was attributed in most cases35 to a complex magnetic spiral, or helimagnetic structure discovered by Sosnowska et al33 using high resolution time-offlight neutron diffraction. The spiral structure is stable at most temperatures below the TC.33 When grown as an epitaxial film, BiFeO3 is usually grown with SrRuO3 electrodes on SrTiO3 single crystal substrates by pulsed laser de position. Research on the variation of strain and thickness effects of epitaxial films of BiFeO3 grown via pulsed laser deposition shows that the polarization does not exhibit a large change in size, despite distinct crystallographic changes

PAGE 25

25 in structure.36, 37 BiFeO3 can be grown on (100) or (111) orie nted single crystal substrates, and has a large polarization of 100-115 C/cm2 in the [111] direction, similar to the bulk.36-38 BiFeO3 is interesting to study because it can be used as a lead-free ferroel ectric and multiferroics material for nanoscale devices.4, 38-41 Multiferroics Introduction There is growing interest in materials that have both magnetic and electronic properties working together into what are termed as multifunc tional materials. In principle, these materials could enable single device components to perf orm multiple functions. Multiferroics that are ferroelectric and ferromagnetic are attractive for the magnetoelectric effect. One proposed application for this additional functionality c ould include a type of electric-field controlled magnetic data storage device.40 Multiferroics are materials that demonstrate two or more ferroic properties (i.e. ferroelectricity, ferromagnetism, and/or ferroelastic ity) in the same phase. There are different types of phenomenon. The magnetoelectric material can exhibit: 1. a spontaneous magnetization that can be reor iented with an applied electric field, 2. a spontaneous polarization that can be reor iented with an applied magnetic field. Coupling these phenomena adds anothe r degree of freedom in the cu rrent device applications in mass storage.40, 42 As mentioned before, multiferroics are materials that exhibit simultaneous ferroelectric and ferromagnetic properties. Magnetoelectric behavior can be dete cted by measuring for example, the magnetoresistance (the effect of a magnetic field on the c onductivity) and/or the magnetocapacitance (the magnetic field dependen ce on the dielectric properties) around the Curie temperature. Multiferroic materials are uncommon in bulk materials. Some possible

PAGE 26

26 restricting factors have been theo retically explored. N. A. Hill3 suggested that d-orbital occupancy of the B-site cation in perovskite ma terials is a critical variable in reducing the propensity for perovskite material s to exhibit ferroelectricity. This consequence reduces the chance for them to be a multiferroic, unless there are extra influences (such as ferroelectricity via lone pair mechanism) that allow for this phenomenon. Single Phase Multiferroics Single-ph ase multiferroics were identified for the first time in 1958. Currently there are more than 80 single-phase multiferroics grown as a solid-solution or a discrete composition. There are four major kinds of single-phase multiferroics as reported by Fiebig:42 1. Perovskite structure materials (ABO3 or A2BBO6) that have slightly deformed cubic symmetry can be multiferroic. One example of this kind of material is BiFeO3, a ferroelectric, ferroelastic, and antiferromagne tic compound. Various solid solutions based on this compound have been made due to intere st in its high electric and magnetic ordering at high temperatures. 2. Hexagonal structure materials that have the same stoichiometric formula as perovskite but crystallize in hexagonal structure can also be multiferroic. These mainly consist of manganites with formula RMnO3 where R = Sc, Y, In, Ho, Er Tm, Yb, or Lu. The point group they possess is 6mm. These compounds are ferroelectric with an antiferromagnetic Mn3+ sublattice. 3. Boracites (M3B7O13X) are ferroelectric, ferroelastic antif erromagnets, that sometimes also have a weak ferromagnetic moment. The range of bivalent ions can be M = Cr, Mn, Fe, Co, Cu, Ni and the anions, X = Cl, Br, or I. These solid solutions have ferroelectric TC > 300 K. However the magnetic transitions occur below 100 K. 4. BaMF4 compounds where M = Mg, Mn, Fe, Co, Ni, Zn, are ferroelectric ferroelastic with solely antiferromagnetic or w eak ferromagnetic ordering. These solid solutions have 2mm point symmetry at high temperatur es and their melting temperature is less than their Curie temperature.3, 42 Although multiferroic phenomena are observed in thes e materials, they are of little utility for applications due to the temper ature range of the coupled pheno mena or the nature of the magnetic ordering (i.e. antiferromagnetic).40

PAGE 27

27 Multiferroic Composites Several groups have examined nanostructured materials as potential multiferroics. For example, self-assembled BaTiO3-CoFe2O4 nanocomposites have been studied by Zheng et al43 and are shown in Figure 2-2. The structures consist of CoFe2O4 nanopillars that are grown perpendicular to a SrTiO3 substrate, with the p illars embedded in a BaTiO3 matrix. This selfassembled matrix is a spinel/perovskite structure grown from a single 0.65BaTiO3-0.35CoFeO4 target by pulsed laser deposition. The author s proposed magnetoelectric coupling through magnetorestriction, a mechanism where there is str ong elastic interactions between the two phases. One of the interac tions comprised of the BaTiO3 matrixs ferroelectric structural distortion from cubic to tetr agonal phase which caused a compressive strain in CoFe2O4 nanopillar along its axis. A drop in th e magnetization at the ferroelectric TC was quoted as evidence for magnetoelectric coupling.43 Multiferroic Materials via Artificial Superlattices Recent m otivation for thin film artificial supe rlattices for multiferroics has increased over the past decade because of the benefit of having ve ry defined interfaces at the atomic level. The interfaces comprised of a magnetic and ferroelectr ic material layered on one another has great advantage over bulk materials because of mani pulation of electron-mediated magnetic dipole ordering. Shown in Figure 2-3, lattice strain effects, dimensionality, and st acking periodicity are some of the key parameters that can be controlled. Tabata et al44 suggested that superlattice stru ctures can realize the following: 1. Strain Effect; lattice stress can be introduced at the interface 2. [Reduced] Dimensionality; first layer is is olated by the second layer with changing thickness of the second layer 3. Stacking periodicity (cycles); it is like a sharp superlatti ce in semiconducting devices. 44

PAGE 28

28 There are two kinds of strain that can occur in a multilayer structure: the interface between the film and the substrate, and the interface between the two distinct materials used to create the multilayers. Thus, one can expect a variation of strains in a multilayer system. In multiferroic systems, compressive (or tensile) strain at th e interface can suppress (or enhance) multiferroic coupling. With reduced dimensionality, the thic kness of the layers can pose variation in the measured property of interest. An enhancement or suppression of coupling can also occur with changing thickness of one or both la yers. Variation of stacking peri odicity can tune the transition temperature of the material property of inte rest, i.e. Curie or Neel temperature. In recent years, superlattices have offered novel approaches to crea te new materials that can be tailored depending on the kind of la yer materials, thickness, morphology, and layer interfacial structure. Varying one or more of these factors does in fact govern the electronic and magnetic properties.45 One of the key roles of synthesizi ng artificial superl attices for use as multiferroic materials is that the properties can yield larger coupling at moderate conditions than single phase multiferroic materials, of which the la tter majority have weak multiferroic coupling. The latest studies of multiferroic material s, especially renewed interest in materials showing simultaneous ferromagnetic and ferroelectr ic properties, are increasingly attempted via thin film superlattice assembly. There are so me groups that have succeeded to assemble ferromagnetic/ferroelectric hetero-epitaxial layers Mentioned here are a few studies in recent years of ferromagnetic/ferroelectric thin films assembled via superlattice technique using laser ablation.46-49 Murugavel et al 48 chose ferromagnetic Pr0.85Ca0.15MnO3 (PCMO) and ferroelectric Ba0.6Sr0.4TiO3 (BST) to fabricate artificial thin film superlattices. They chose PCMO, because it appears to be ferromagnetic and insulating which is uncommon, and BST, for a minimum lattice

PAGE 29

29 mismatch with PCMO. They observed increas e in magnetoresistance in the samples with increasing ferroelectric layers as well as magneti zation and dielectric cons tant as a function of temperature, indicating that superlattices could have coexistence of ferromagnetic and ferroelectric properties.48 Another group50 created multilayers of ferroelectric BaTiO3 (BTO) and ferromagnetic La0.7Ca0.3MnO3 (LCMO). They varied the thickness of the BTO layers while keeping LCMO layers constant at a thin layer where it behaves as an insulating ferromagnet. They revealed that the total magnetization increased as the BTO la yer thickness is increased. However the Curie temperature of the films were almost independent of BTO thickness. In addition, they obtained successful measurements of hi gh magnetocapacitance and large ma gnetoresistance, which they claim will open the path to design optimal multiferroic thin films.50 Multiferroic Solid-Solution Epitaxial Films There has been m uch work on enhancing the magnetic propert ies of single phase multiferroic materials that exist, for example by substitution and doping of A-site and/or B-site cations.51-71 BiFeO3based solid solutions have recently become of increasing interest, as research groups look to make this antiferroma gnetic ferroelectric exhi bit maximum strength magnetism without compromising the integrity of its robust ferroelectric properties.54, 56-61 Only a few groups have investigated the properties of Aand/or B-site cation substitution on bulk BiFeO3,54-59, 62-64 and even less articles are published on Aand/or B-site cation substitution in BiFeO3 epitaxial thin films.60, 61, 70, 71 A-site cation substitution is of importance in BiFeO3 because it has been theoretically predicted and experimentally show n that the doping or substitution of certain elements can cause a change in the proposed ma gnetic structure of BiFeO3.72, 73 However, this is not an easy task, taking note that even the s ynthesis of single phase BiFeO3 in bulk or thin film is rather difficult

PAGE 30

30 to achieve due to many factors, including the vo latility of Bi ions, narr ow window of epitaxial deposition constraints, and nu merous possible impurity energetically favorable phases.31, 35, 74 Hence, another reason to utilize substitution into BiFeO3 is by adding another solid solution in the system that can prevent the formation of second phases and improve the resistivity of BiFeO3.32, 51, 55, 75 Some groups have utilized isovalent cations, ma ny which are rare earth elements such as Gd, Nd, Sc, La, and Ce, as an A-site substitution.54-61 Some have substituted as little as 5%, and at most 40% substitution for Bi. Khomchenko et al showed that with adding up to 30% Gd to bulk BiFeO3, the room temperature magnetization increased to weak ferromagnetism. They claim it was not due to Gd3+ ions themselves being magnetic, but that an antisymmetric exchange mechanism was the reason for the increase, as we ll as the main idea that there is a substitutioninduced suppression of the spiral spin modulation.54 Two groups studied the effects of substitution of Lanthanum (La3+) into bulk BiFeO3. Zhang et al claim that adding La3+ induced the destruction of the spin cycloid, and hence resulted in a magnetizati on two times that of pure BiFeO3. The second group, Das et al, demonstrated that the presence of a small secondary phase in BiFeO3 was removed upon La substitution at the Bi site, which meant that stabilization of the bulk crystal structure wa s enabled. In addition, they concluded that nonuniformity caused by La substitution enhanced the multiferroic properties of BiFeO3.56 As mentioned before, there are only a few articles published for heterovalent A-site substitution in BiFeO3 in the form of thin films,60, 61, 70, 71 which is not a surprise due to the relatively new general interest in improving multiferroic material s by cation substitution, rather than just by growth conditions a nd substrate choice. Epitaxial f ilms that are grown, especially via pulsed laser deposition, discusse d in Chapter 3, have proven to be relatively advantageous.

PAGE 31

31 This is due to the ease at which most any materi al can be grown with stoichiometric transfer of atoms onto a single crystal substr ate. Two groups in particular have grown the same A-site substituted material, (Bi0.6Tb0.3La0.1)FeO3, both via pulsed laser deposition, and a third research group synthesized this mate rial as a bulk ceramic.63, 70, 71 The findings of Palkar et al of bulk (Bi0.825Tb0.075La0.1)FeO3 were a decrease in the ferroelectri c transition temperature (they ascribe this to the addition of Tb3+ which causes a decrease in th e unit cell volume), and a room temperature magnetization increase to 4B per Tb ion.63 This is the largest magnetization reported for a multiferroic material. A second group, Wang et al, grew epitaxial films of (Bi0.6Tb0.3La0.1)FeO3 and reported that growing at very low pressures removes the presence of impurity phases (an BiFeO3 orthorhombic phase and a ferric ox ide) and increases the crystal linity, in addition to decreasing the leakage current at electri c fields higher than 50 kV/cm.72 The authors believe the cause of leakage current is attribut ed to a small amount of Fe2+ ions or oxygen vacancies. They concluded the cause was the presence of oxygen vacan cies and reasoned their findings in a paper where doping BiFeO3 with Ti4+ ions decreased the leakage in those films, indicating that oxygen vacancies was the main cause.76 The authors also ex situ annealed the films in oxygen ambient atmosphere and measured the lowest leakage curren t ever reported in literature on pure or doped BiFeO3 thin films.72 The third and last report on (Bi0.6Tb0.3La0.1)FeO3 served as a response to the two previously mentioned articles The authors, Eerenstein et al,70 attempted to reproduce data produced by the previously mentioned articles a nd explained their own observations. First, the authors addressed the problem of la rge variation of measured BiFeO3 polarization (between 2 and 150 C/cm2), especially groups measuring polarizations in excess of 100 C/cm2. They state

PAGE 32

32 that to determine if a measur ed polarization is valid, freque ncy-dependent measurements and leakage density (acceptable values are below 10-5 A/cm2) should be reported with polarization measurements. In addition to this concern, Eerenstein et al challenged the claim of a large ferromagnetic moment of 4B per ion.63, 70 They reproduced result s where the magnetization increases from 0.05B per unit cell in undoped BiFeO3 films to ~0.08B per unit cell for (Bi0.6Tb0.3La0.1)FeO3. The concluded that it is possible th at the large value reported was due to iron oxide impurities due to Bi loss. They also added that the large room temperature moment cannot be explained by A-site magnetic order, becau se this is only expected as low temperatures (< 5 K), and other ra re earth substitutions54-61 show only weak magnetism.70 To conclude, there is much investigation a nd interest in A-site substitution of BiFeO3 among other methods of finding ways to improve on the setbacks of the already complex properties of this multiferroic material. Why choose BiFeO3 to combine with BaFeO3-x? Because BiFeO3 is one of the most promising lead-fr ee piezoelectric materials that exhibit multiferroic behavior at and above room temperat ure. Part of this work is about ways to combine these two materials in a solid solution to study their properties, in particular their structural and magnetic properties.

PAGE 33

33 Figure 2-1. Schematic of a perovskite stru cture, below and above the ferroelectric TC. Some ferroelectrics, such as this, the B cations shift off position.

PAGE 34

34 Table 2-1. Single phase multiferroics with various properties. Compound Structure Point Group(s) Properties Example Perovskite ABO3 or A2BBO6 3m FE, FEL, (A)FM BiFeO3, PbFe0.5Nb0.5O3 Hexagonal Manganites RMnO3 6mm FE, AFM R = Sc, Y, Ho, Tm, In Boracite MB7O13X High T: bar43m Low T: 3m or m FE, FEL, (A)FM M =Ni, Co, Cr X =Cl, Br, I BaMF4 BaMF4 2mm FE, FEL, (A)FM M = Mg, Mn, Fe

PAGE 35

35 Figure 2-2. Schematic of self-assembled nanostr ucture thin film aligned perpendicular to substrate. Reprinted with permi ssion: H. Zheng, et al., Science 303, 661 (2004). Figure 2-3. Variation of layering in an artificial superl attice. A) Strain: Lattice stress can be introduced at the interface; B) Dimensional ity: black layer is isolated by the white layer with changing thickness of the white layer; C) Stack ing: change in the number of cycles. Reprinted with permission: H. Tabata, K. Ueda, and T. Kawai, Materials Science and Engineering B 56, 140 (1998).

PAGE 36

36 CHAPTER 3 METHODOLOGY Synthesis of Thin Films Pulsed Laser Deposition Introduction Pulsed Laser Deposition (PLD) is a thin f ilm growth technique that is conceptually simple. The apparatus consists of a target hold er and substrate holder located inside a vacuum chamber. A laser is used to ablate materials fr om the target and deposit on substrate(s) mounted to the substrate holder. Film growth can be ca rried out in almost any kind of gas environment with or without plasma excitation.77 A brief history, sp ecific equipment used, and technique of PLD will be briefly explained here. After the first high-power ruby laser be came available, many experimental and theoretical studies of interacti ons of intense laser beams on solid surfaces, liquids and gaseous materials came about.77, 78 However, there were not many research studies from this time in the 1960s to the mid 1970s, when the electronic Qswitch was developed, which could send short pulses with high power intensity, with the help of a high efficiency second harmonic generator, which reduced the splashing and increased abso rption depth. This increased the variety of materials that could be studied with lasers, but it wasnt until the breakthrough of epitaxial growth of High TC superconducting films in 1987 that PLD broke out as a high interest growth technique of choice.77-79 Why pulsed laser deposition? One of the greatest advantages of PLD is its capability to explore almost any combination of materials (both undiscovered and currently st udied) compared to most epitaxial growth systems. Unlike other systems, the energy source of PLD, the laser, is independent from the

PAGE 37

37 deposition chamber system, making PLD a simpler technique to create co mplex epitaxial films.77 In addition, the ability to realize stoichiometric transfer of ablated material from multi-cation targets shows that almost any material that exis ts in the bulk form can be studied via growth by PLD as an epitaxial thin film.78, 80 There are a large variety of c onditions that can be varied via laser mirror optics, target-substrate distances, am bient gases, and various laser energies. Components Laser Laser wavelength is im portant in the use of PLD. To effici ently ablate target material, temperatures of the ablated volume must be well above what is required for evaporation of species. Thus the laser pulse should have shor t duration, be high in en ergy density, and highly absorbable by target materials. Specifically for ceramic target pellets, ultraviolet wavelength lasers are optimal for achievable deposition.78 Generally, laser wave lengths for PLD exist between 200 and 400nm, Table 3-1 shows different excimer laser operating wavelengths. These wavelengths are used because most materials exhi bit strong absorption in this spectral region. Most research has been executed with excimer and Nd3+:YAG (Yittrium Aluminum Garnet) lasers as the energy source. Light output from an excimer laser comes from a molecular gain medium, and the lasing action occurs between an upper electronic state and a weakly bound ground state, and it emits a photon during the transition from one to the next. For a high-quality deposition, the laser output must have uniformity ; hot spots and non-uniformity must be avoided, especially in the use of multi-component depositi on targets. Non-uniformity in laser output can lead to particle formation on substrate, or non-stoichiometry in the epitaxial film. KrF, having the highest gain, is one of two of the most extensively used ex cimers for PLD (the other being XeCl).77 In this work, a Lambda Physik Comp ex 205 KrF excimer laser was utilized for

PAGE 38

38 epitaxial films grown in Chapter 4, and an LPX 325i KrF excimer laser was used for epitaxial films grown in Chapters 5 and 6. Laser-target interaction A laser puls e from the KrF-excimer laser strikes the target, and a fraction of the energy is reflected and the rest is absorb ed by the target. During the lase r pulse, heat conduction occurs at targets surface area and certain depth length. This makes a heated volume with energy inside, where solid material within the volume become s highly dense and genera tes into plasma. Any leftover energy causes bonds to break and material is ablated from the surface of the target into a plume. The effusion layer (plume) expands out of this shallow target crater, mostly perpendicular to the surface. The effusion layer cools during the adiabatic expansion and at the same instant the atoms acquire a large flow velo city in the surface norma l direction towards the heated (and in some cases rotating) substrate.81 Optics The optics between the laser output and the vacuum cha mber laser viewport consist of beam splitters, apertures and focusing lenses, laser window, and mirrors. It is incredibly important that the optics remain clean and defect free, for dirt or dust particles that land on the optics can be permanently embedded in the lenses causing major damage. Spherical lenses can magnify in two orthogonal planes, which image a poi nt source as a point, as well as change the position of the image in up or down and left or ri ght. Cylindrical lenses magnify in one plane, and so they change the position of the image in only one direction.77 Deposition chamber The basic PLD deposition system consists of th ese parts: chamber, target manipulation, substrate heater and holder, gas flow controllers, pressure gauge s, differential vacuum pumps, and various gate valves. Some syst ems also include extra apparatus for in situ growth

PAGE 39

39 measurements including Refl ection High Energy Electron Di ffraction (RHEED), ion and elemental probes. Figure 3-1 is a schematic of a Laser Molecular Beam Epitaxy (LMBE) system used for the growth of films in Chapter 4. Targets BaFeO3 targets were synthesized from BaCO3 (99.997%) and Fe2O3 (99.998%) powders that are crushed and dry-mixed in a high density alumina mortar and pestle set for 45 minutes. The powder mixture was placed in a covered high density alumina boat and calcined in a box furnace in air in the following sequence: )(25 950 950 800 )(5 6 2 4RTC C C C RT RoomTemphrs hrs hrs hrs After removal, the powder mixture was put in a ball mill and mixed dry for 30 minutes, or dry-mixed a second time with alumina mortar and pe stle set. Then the mixed powders were cold pressed into a 1 diameter pell et with a stainless steel die and held under 2000 lbs psi for 2 minutes and then placed again in a box furnace in ai r and set to sinter in the following sequence: )(25 1000 1000 )(5 12 4RTC C C RT RoomTemphrs hrs hrs The target is then checked for hardness and polishe d with sand paper to ensure a flat surface for even laser ablation. (Bi, Ba)FeO3 solid solution targets were synthesized using the same conditions as BaFeO3 targets, but with an extra step in the process where Bi2O3 was added right after the BaFeO3 mixture was calcined. Due to the vol atility of Bismuth, the calcining and sintering temperatures were 100 degrees less than that of BaFeO3. In addition, 10% excess Bi2O3 was added to account for Bi lo ss during sintering and deposition.78 Three targets: SmScO3, SrTiO3, and KTaO3 targets were obtained from Darrell Scholms research group (Penn State), Praxair, Inc, and ORNL (hot-pressed).

PAGE 40

40 Growth procedure The ceram ic target is loaded into the depositi on chamber on a target carousel. The laser is used to ablate the targets. The pulsed laser is focused through a series of mirrors and lenses mentioned previously, eventually through th e deposition chamber window onto the rotating target.82 Film deposition occurs on the heated (and in some cases rotating) substrate. The substrates utilized in this work are SrTiO3 (100), LaAlO3 (100)pc KTaO3 (100), NdGaO3 (100)pc, DyScO3 (100)pc, and (La0.29,Sr0.71)(Al0.65,Ta0.35)O3 (LSAT) (100)pc single crystal substrates. Although LaAlO3 is rhombohedral at room temperature, it is a cubic perovskite at the growth temperatures and can be indexed as a perovskite pseudo-cubic with a0 = 3.789 at room temperature. Prior to gr owth, the substrates were cleaned with trichloroethylene, acetone, and methanol in an ultrasonic bath, followed by drying with dry nitrogen. In order to remove any trace metals that might introduce spurious magnetic properties, the unpolished backside of the substrate wa s etched with a 50/50 solution of HNO3 and distilled water. The base pressure of the deposition is sustained at 10-7-10-9 Torr, with various growth pressures maintained between 0.1-300 mTorr Po2. Oxygen and Nitrous Oxide flow rates are 1-2 sccm. The temperatures of the substrates m ounted on substrate holders during growth are 600900C respectively. Laser ablation energies are between 2-3 J/cm2 with a pulse rate of 5 or 10 Hz for BaFeO3 epitaxial films, and 1.68 J/cm2 at 50 Hz for continuous-target-rotation mechanism for solid solution thin films. Two -target continuous rotation method Epitaxial film s discussed in Chapter 6 were grown using a two-target continuous rotation method. This method is used to create solid solution or alloy epitaxial films via the use of two targets, instead of making a ne w target of the desired composition. Before deposition, the

PAGE 41

41 desired solid solution composition is calculated, to know how many shots per end-member target is needed, and how many repetiti ons are required for the desired thickness. This information is put into a text file, and then loaded into a Labvi ew software program whic h will convert the text file to a run sequence. In the deposition chamber, two targets are placed opposite from each other on the target carousel, a nd routine preparations are comp leted prior to deposition. The program, when executed will command the entire targ et carousel to rotate at a certain speed, and laser pulses will strategically hit each target with a certain number of pulses at determined frequencies to create alloy mixing. This method, when modified, can also be used to create artificial superlattices. Characterization Superconducting Quantum Interference Device Magnetometry (SQUID) Introduction The m agnetic properties of the samples ar e investigated by using Superconducting Quantum Interference Device (SQUID) Magnetometry. The SQUID consists of a superconducting ring with an insulating weak lin k (a.k.a. Josephson junction). These junctions allow discrete increments in flux to enter the ring The changes in flux are detected by a voltage induced in a coil placed around the superconducti ng ring. These counts or voltage pulses are used to measure the applied field.83 Magnetic properties can be determined by finding how the magnetization changes versus temperature, as well as with applied field. Theory To further explain the Josephson junction, it is important to briefly note the Josephson Effect, a phenom enon predicted by Josephson as a gr aduate student at Cambridge in 1962. The junction consists of a thin insulati ng film separating two superconductors.6, 8, 83 Cooper pairs (a pair of electrons that have a lower energy than the two individual elect rons) can tunnel through

PAGE 42

42 this barrier, but only up to a certain critical current which changes by the presence of a very small field. SQUID uses this change in curr ent to detect small changes in magnetic field.7, 8, 83, 84 Magnetic Property Measurement System (MPMS) comp onents When speaking about what a SQUID is, it mu st be acknowledged that a SQUID is just one of four main superconducting components th at are located inside a Magnetic Property Measurement System (MPMS) that is manufactur ed by Quantum Design. Please note that there are many system components in the MPMS, how ever only the superconducting components will be explained. The first superconducting component, the SQUID is an incredibly sensitive device that doesnt directly detect the magnetic field from a samp le, but when the sample is moving through the system of superconduc ting detection coils (another s uperconducting component that are connected to the wires of the SQUID devi ce), the current from th e detection coils is converted to voltage through the sample and the output voltage is recorded. The SQUID therefore functions as a cu rrent to voltage converter.84 More importantly, the basic function of the SQUID is measuring magnetic dipoles (or magnetization). When a measurement is performed in th e MPMS, a sample is moved through the superconducting detection coils that are located outside the sample chamber and at the center of the superconducting magnet (a third superconducti ng component). The magnet consists of a solenoid shape, where it is constructed as a closed superconducting loop. The fourth component is the superconducting shield, which is an equally important component as the others due to the sensitivity of the SQUID detector. The magnetic field produced in the SQUID by a small sample such as a thin film is on the order of a fl ux quantumwhich when compared to the magnetic flux though a 1 square centimeter of the earths magnetic field, it is 9 orders of magnitude larger than the samples.84

PAGE 43

43 Sample preparation Silver p aint is used to attach the samples to the substrate holder during film growth. The silver paint must be removed so that there is no magnetic moment from impurities in the silver paint. Thus there will be no extra work to scan a reference sample with si lver paint and have to worry about subtracting the silver paint moment from the sample moment. Most importantly, the magnetization data can be counted as real, since the epitaxial films magnetic moments are relatively weak, compared to bulk samples due to orders of magnitude less mass. There were two ways silver paint was removed from samples. The first way remove the silver paint from the sample, a layer of photo-resi st is spun onto the film surface of the sample, heated on a hotplate to be cured. The photoresist is used to protect the film layer of the sample from damage due to handling and from silver dust particles. After the minute is completed the sample is cooled by placing it on a rectangular meta l block to pull the heat from the sample. After five minutes, when the sample is complete ly cooled, the back a nd sides of the sample are sanded with ultra fine sandpaper. After checki ng that all traces of silver paint are sanded off, the sample is then rinsed with deionized (DI) water, and air-dried with nitroge n gas. This step is to ensure there are no loose silv er paint particles on the sample. To remove the photoresist, the sample is rinsed with Acetone solvent, then Isopropanol alcohol, and then air-dried with Nitrogen gas. The sample is checked under a ligh t microscope for traces of photoresist residue and/or silver paint. The second way to remove silver paint is to do the same process as mentioned above, but instead of sanding the back of the sample with sandpaper, the back can be etched with a 50/50 solution of HNO3 and distilled water. Sample mount To prepare a sam ple to mount in the MPMS, the thin film samples are placed in plastic straws. Kapton tape, which is used as a low b ackground noise tape, is then placed on one end of

PAGE 44

44 the strawset to prevent the samp le from slipping out the sample mount strawset. The samplemount strawset is now ready to attach ont o the MPMS metal sample transport rod. The sample-mount strawset open end is slipped onto the MPMS sample transport rod, and subsequently secured with more kapton tape The sample transpor t rod with attached sample-mount strawset is placed in the airlock of the MPMS chamber and pushed down into the MPMS. Usually the temperatur e inside the MPMS hovers around 30 Kelvin, so the sample is cooled quickly to that temperature when pushed deep into the MPMS. Sample measurement Once inside the MPMS c hamber, the sample is re ady to be centered. First a low field is set to be used to center the sample. In the expe riments executed in this work, a low field of 2000 Oersted (Oe) was used. The temperature to center the sample can be any, but in these experiments, usually 40 Kelvin, or a temperat ure well below the samples guessed or known Curie Temperature (TC) was used. The top of the sample transport rod is attached to a stepper-motor-controlled platform that is used to drive the sample th rough the superconducting detection coils in a series of defined steps. During the centering process, the sample is stopped at several positions over a specified length, here four centimeters, and several read ings of the SQUID voltage are recorded and averaged. Depending on whether there is a nonmagnetic (actually diamagnetic or paramagnetic) response or a magnetic (actuall y ferroor antiferromagne tic) response, that maximum or minimum voltage at a certain scan distance (in centimeters) will be where the sample is located. If the voltage versus scan di stance is inconclusive, then a higher field can be applied, or restart the process of mounting the sample (to check if the sample was not secured properly).84

PAGE 45

45 After the sample is centered, the field can be turned off,, and the saved sequence files can be executed. There are 2 main kinds of scans that are executed in these projects: 1. M(H): Magnetization as a func tion of applied magnetic field 2. M(T): Magnetization as a function of temperature H is the magnetic field applied to the samp le by the superconducting coil, and T is the temperature applied. M is the magnetization (or ma gnetic response) of the sample. In an M(H) measurement, measurement is taken at a fixed te mperature (in these projects it is 10 Kelvin) while measuring M at a series of H values. In an M(T) measurement, the applied field is fixed (in these projects the field was 2000 Oe usually) while measuring M at a series of temperatures. In these projects there were tw o kinds of M(T) measurements: field-cooled and zero-field cooled. In the field-cooled M(T) measurements, the sample is taken to room temperature (300 Kelvin) and then cooled to about 5 Kelvin in a low field (2000 Oe) while measuring M. In the zero-field-cooled M(T) measuremen t, the sample is cooled without a field to about 5 Kelvin, and then raised to room temperature in a low field (2000 Oe) while measuring M. These two temperature dependent measurements are one way of determining if irreversibility exists in the sample. The samples were measured in a Quantum Design SQUID by Dr. Hebards research group in the Physics Department at University of Florida as well as by the author in the Materials Science and Technology Division at Oak Ridge National Laboratory. To accurately measure the magnetization of the film, substrates were measur ed with SQUID, and their diamagnetic (or in some cases paramagnetic) response was subtract ed as the background from the actual samples measurement. This was accomplished by taki ng the magnetization versus field data, and dividing this by the mass of the reference substrate. Then these values were subtracted from the sample data, and the data was multiplied by the samples actual mass. All samples in Chapter 4

PAGE 46

46 were measured with the applied field perpendicular to the plane, or [001] direction. Most samples in Chapters 5 and 6 were measured with the applied field parallel to the plane, or [100] direction. In some cases samples in Chapter 5 were measured perpendicular to the plane. In Chapter 6, the hysteresis loops are examine d, and what is noteworthy is that there is paramagnetic signal coming from all four samples, which is attributed from the substrates. LAO and STO are mostly diamagnetic, so their bac kgrounds are relatively easier to subtract. However, LSAT and NGO are strongly paramagnetic making it is more difficult to subtract a large paramagnetic component in addition to a diamagnetic background. This paramagnetic response from a substrate, which can be tens to thousands of times greater than the films magnetic response, greatly distorts what would look like normal ferromagnetic hysteresis loops in these samples. X-Ray Diffraction (XRD) Introduction X-ray Diffraction (XRD) is a technique used to m easure the structural properties of a material such as strain, epitaxy, phase composition, preferred orientation, and defect structure. XRD is nondestructive and can be used in mo st environments, making it advantageous over other techniques that can also determine crystalline phases such as Transmission Electron Microscopy (TEM) which is a de structive technique. XRD can be used to determine the thickness of thin films and multilayers. It is important in many technological applications because of its ability to determine strain states. For magnetic thin films, it can be used to uniquely identify phases and pr eferred orientations, since these can determine magnetic properties.83

PAGE 47

47 Braggs Law (X-ray Theory) Crystals are m ade of planes of atoms that ar e spaced a distance, d, apart, each family of planes has a different d-spacing fr om one another. To differentia te these planes, a coordinate system, such as an orthogonal system for a simple cubic crystal can be un iquely distinguished by its Miller Indices. The Miller Indices with the a, b, and c crystallographic axes, are the reciprocal intercepts of the plane reduced to the smallest integers having the same ratio.85 An (hkl) plane intercepts the axes at a/h, b/k, and c/l. Thus the dspacing between (hkl) planes is dhkl for simple cubic crystals, where 222 0lkh a dhkl ,..(3-1) Here in Eq. 3-1, a0 is the lattice constant of the crystal.85 In X-ray diffraction, there are three basic pa rts (probe, interacti on, detection) that contribute to observing diffraction in a crystal. The probe is the incident x-ray beam, which is focused with measured slits to reduce beam size. The beam interacts with a specimen which, in this project is an epitaxial thin film on a singl e crystal substrate. Under certain condition, there will be constructive interference which enables the incident beam to diffract off the crystals atomic planes. The detection is the detector which the diffracted beam (x-ray photons) enters into and is recorded as a specific d-spacing. Braggs law gives the condition for which a diffraction peak can be observed from the cons tructive interference from x-rays scattered by atomic planes in a crystal: hkl hkld sin2,....(3-2) In Eq. 3-2, hkl represents the angle that is between th e planes and the incident and diffracted xray beam. To observe a diffracted plane, the detector must be positioned so the diffraction is 2hkl. In addition, the crystal should be oriented so that the norm al to the diffracting plane is

PAGE 48

48 coplanar with the incident and diffracted beams, so that the angle between them is equal to the Bragg angle hkl. Only one specimen orientation is possible for each (hkl) plane in single crystals and epitaxial films.83, 85 Reciprocal space maps Reciprocal Space Maps (RSMs) are two-axes m easurem ents that produce a twoor threedimensional map of the epitaxial film and substr ate, mostly with off-axis planes. These scans are incredibly useful because ca lculation of the out in-plane as well as out-of-plane lattice parameters can be obtained. The measurements are loaded into X-Pert software. Using the substrate as the reference point, the peak distance differences (or mismatch) from the substrate to the film is measured. The parallel mismatch the difference taken from the x-axis and perpendicular mismatch is taken from the differen ce of the y-axis. For more accuracy, take into consideration the average of the K 1 and K 2 peaks. These differences can be converted to lattice constants when the lattice constant of the substr ate is known. On reciprocal space maps one can identify whether a film exhibits strain or relaxation and to what degree, with respect to the substrate. It is important to note that in the epitaxial films measured here, it assumed that the films have some sort of tetragonal stru cture when calculating lattice parameters. Other scans Two-theta (2) scans can determine the unique d-sp acings of crystalline materials and specifically show out of plane p eak intensities, which is proof of epitaxy. In this project specifically, the perovskite ma terials studied will show (00l) peak intensities. Omega ( ) Rocking Curve scans will oscillate about a chosen d-spacing on the 2-th eta/2 axis (or omega) and the information determined is the crystallinity of the sample. The breadth of the rocking curve or Full-width-at-half-maximum (FWHM) w ill determine the distribution of the crystalline grains. A narrow rocking curve (FWHM 0.5) means the film is highly crystalline, whereas a

PAGE 49

49 FWHM greater than 0.5 is poorer crystalline qua lity. Single crystal substrates always yield a FWHM of less than 0.08. Phi scans verify or identify the symmetry of the material measured. In addition, specific families of planes of single crystal subs trates can be determined. Components used in the study In this work, a couple of diffractom eters were used: a Phillips APD 3720 two-circle x-ray diffractometer for films grown in Chapter 4, and a Pan Alytical four-circle x-ray diffractometer for films grown in Chapters 5 and 6. In the f our-circle diffractometer, a mirror was used (which limits peaks to the presence of K 1 and K 2 peaks. Also slits of and were used in measurements. Scanning Electron Microscopy (SEM) Scanning electron microscopy operation A scanning electron m icroscope (SEM) obtains topographic images and elemental analysis of organic and inorganic materials. Its usefulness stems fr om its capability of attaining threedimensional-like images of the surfaces of various samples and specimens. It has a magnification range of 10-10,000, and most elemen ts can be identified using the electron dispersive spectrometer portion of the SEM. Th e SEMs instrumental resolution is generally on the order of 10-50 Angstroms; as well it has a larg e depth of field, which is responsible for the three-dimensional appearances of sample imaging. Overall the SEMs most important use is for structure analysis and elemental analysis.86 The SEM is made of two major components, an electron column and a control console. The electron column contains an electron gun and two or more electron lenses, which mostly control the path of the electrons travelling down the column. The control console is made of a cathode ray tube (CRT) viewing sc reen and computer keyboard with knobs that are both used to control the electron beam.86

PAGE 50

50 To describe how the SEM works, there are basic stages that can be expounded upon: the electron gun and lenses produce an electron beam, a deflection system controls the magnification, the electron detector collects the signal, and the co mputer records the image. First the electron gun generates electrons (usually by work of an applied voltage across a heated tungsten filament) which are then demagnified by electron lenses, because the spot size is too large. The electrons are focused in a smaller be am, and this beam goes through lens at the end of the column into the sample chamber, where it in teracts with the sample to a depth of around 1 micron. In the deflection system, two pairs of el ectromagnetic deflection coils are used to sweep the beam across the sample, which creates a raster on the viewing screen. Then distinction or contrast in the image occurs during beam-sample varied interactions that are collected by the electron detector, which convert s the signal to intensity chan ges on the viewing screen and produces an image. Last, this image is stored in digital form by a computer for processing and printing.86 Sample preparation and mounting Due to bombardm ent of elec trons onto the sample during SEM, if the sample is nonconducting, (or insulating) the sample may e xhibit a phenomena called charging. To overcome this problem, samples can be prepared by coating them. In this work, samples were coated with Carbon, using an SPI Sputter Coater. X-Ray spectral measurement via elect ron disp ersive spectrometer (EDS) Electron Dispersive X-Ray Spectrometers are solid state detectors that are used for x-ray measurement to determine chemical analysis of sa mples. When the electron beam interacts with the sample, x-ray photons from the sample pass through a thin window into an cooled reversebias p-i-n (p-type, intrinsic, ntype) Si(Li) crystal. A photoel ectron is ejected after each x-ray photon is absorbed, which gives up most of its ener gy to form electron-hole pairs. These pairs

PAGE 51

51 are taken by an applied bias to form a charge pu lse, which converts to a voltage pulse, and then the signal is passed to a computer x-ray analyzer. The data is s hown as histogram of intensity by voltage.86 In this work, ceramic target powder samples were examined using a Hitachi S4800 FEG SEM, located in the High Temperature Materi als Laboratory (HTML) at Oak Ridge National Laboratory. As stated earlier, most target powder samples were coated with carbon, and their chemical composition and surface were analyzed.

PAGE 52

52 Figure 3-1. Schematic of Laser MBE Growth Chamber Setup Table 3-1. Excimer Lase r Operating Wavelengths Excimer Wavelength (nm) F2 157 ArF 193 KrCl 222 KrF 248 XeCl 308 XeF 351

PAGE 53

53 CHAPTER 4 FERROMAGNETISM IN PSEUDO-CUBIC BaFeO3 EPITAXIAL FILMS Introduction Magnetic oxide m aterials87 have attracted considerable at tention due to interest in the fundamental science of magnetic moment interactions, magnetic device applications, and the possibility of creating novel coupling via th e formation of nanostructured interfaces.43, 88, 89 One member of the perovskite oxide family that exhibits unique magnetic properties is BaFeO3. It is one of the few oxides where iron assumes an oxidation state of +4.16 In bulk, BaFeO3 normally assumes a hexagonal crystal structure (a=0.568 nm and c=1.386 nm), although various polymorphs are observed with oxygen deficiency.16, 17, 19, 90, 91 The magnetic properties of BaFeO3 are strongly dependent on the oxygen content, presumably affecting the formal valence of the iron cation. Several studies indicate that bulk hexagonal BaFeO3 exhibits an antiferromagnetic to ferromagnetic transition at 160 K, with a ferromagnetic Curie temperature of 250 K.18 More recent Mssbauer and high magnetic field studies seem to contradict this conclusion, showing that no magnetic or dering occurs for 160 K > T > 220 K.19 This discrepancy could be relate d to the different oxygen st oichiometries that BaFeO3-x can assume.18 Recent results for epitaxial BaFeO3 thin films indicate that the structural, magnetic and electrical properties of films can be signifi cantly different from that of bulk material.20, 22, 92 For epitaxy on (001) SrTiO3, BaFeO3 films assume a metastable pseudo-cubic crystal structure similar to that observed for SrFeO3, a material that exhibits antiferromagnetic characteristics with screw spin type magnetic structure and is meta llic. This stabilization of a metastable phase via epitaxial growth has been observed for other perovskite-like oxides.93, 94 Previous reports on pseudo-cubic BaFeO3 films suggest that this phase is ferromagnetic at room temperature,22, 92 in contrast to the hexagonal bulk pha se which is not. This modifi cation of magnetic properties is

PAGE 54

54 presumably the result of strain a nd/or defects in the films. M odification through strain has been observed in a variety of per ovskite material properties.78, 95-97 However, it is unclear how strain might yield an enhancement in the Tc for magnetic ordering in the Fe+4-O-Fe+4 system. The reported lattice constant of the BaFeO3 epitaxial film on SrTiO3 was a=4.120 which is larger than that expected based on anion-cation spacing in bulk materials.20, 22 Note also that the films were also reported to be highl y insulating. At 300 K, magnetization measurements showed hysteresis as well as remnant magnetization.22, 92 In this work, the properties of epitaxial pseudo-cubic BaFeO3-x films grown by pulsed laser deposition are examined. The focus of this work is on the variation of magnetic and structural properties as a function of proce ssing conditions that yield variati ons in oxygen content. As will be seen, the onset of robust ferromagnetism is a strong function of oxygen content, similar to what is seen in the bulk phases. Experimentation BaFeO3 thin films were grown on (012) oriented LaAlO3 and (001) SrTiO3 single crystal substrates via pulsed laser deposition.79, 98 The LaAlO3 substrates were pre-screened in a Superconducting Quantum Interference Device (S QUID) magnetometer to insure that the substrates were non-magnetic (no ferromagnetic impurities) and thus suitable for magnetic characterization of the thin films after growth. Deposition prepara tion is discussed in Chapter 3. Laser energy densities on the order of 2-3 J/cm2 were utilized. Target to substrate distance was on the order of 4 cm. The ablation ta rgets were 1 inch diameter BaFeO3-x ceramics. The thin films were grown at substrat e temperatures between 650-850 C and an oxygen pressure ranging from 0.1-100mTorr. Growth time was 60 or 100 minutes. After growth, the samples were cooled at 10 /min in the oxygen partial pressure used during growth.

PAGE 55

55 The crystal structure and orientation of th e deposited films were examined by X-ray diffraction (XRD). Magnetic ch aracterization was performed on a SQUID magnetometer, with substrate and sample preparation discussed in Chap ter 3, as well as details of what sample crystal orientations the magnetic fields were applied to. In some cases, BaFeO3 films were annealed in oxygen in a high temperature tube furnace. Results and Discussion As-Deposited BaFeO3-x Epitaxial Films Initial studies focused on understanding epitaxy and phase evolution fo r the in situ grown BaFeO3-x films. Figure 4-1 shows the X-ray di ffraction data for films grown on LaAlO3 at temperatures ranging from 650-850C in an oxygen pr essure of 10 mTorr. In all cases, the growth of BaFeO3-x in a pseudo-cubic perovskite struct ure is observed, although not as a single cubic phase. In particular, the diffraction data show the emergen ce of two distinct phases. At low temperatures, the BaFeO3-x (200) peak at a 2-theta angle of 44.27 corresponds to a pseudocubic lattice parameter of 4.09. As the substr ate growth temperature is increased, a second peak emerges at 45.13 corresponding to a latt ice spacing of 4.02. This could represent the emergence of a second cubic phase with a slight ly smaller d-spacing, or a splitting of the peak due to a single, low symmetry phase with diffe rent oxygen stoichiometry. Low symmetry phases are seen for SrFeOx (2.5 x 3.0), where there is a cubic (SrFeO2.97), tetragonal (SrFeO2.86), and orthorhombic (SrFeO2.73) progression with change s in oxygen stoichiometry.99 However, additional in-plane characteri zation of the crystal structure suggests that the BaFeO3-x films consist of two pseudo-cubic phases with slightly different lattice spac ing. This result is consistent with that shown in other work on BaFeO3-x films where shifts in the pseudo-cubic (00l) peaks have been reported.18, 92 The as-grown BaFeO3-x thin films grown at 10 mTorr were translucent with an orange-tan color at a film thickness of a pproximately 300 nm. Transport

PAGE 56

56 measurements show that the as-grown films were non-conducting. The magnetic properties of the as-deposited BaFeO3-x film were measured at temperat ures down to 10K. The magnetization versus applied magnetic field film behavior is shown in Figure 4-2. A small but discernable hysteresis is seen in the M vs. H plot at 10 K, indicating that these films are weakly ferromagnetic. Ex situ anneal treatment of BaFeO3-x Epitaxial Films To investigate the progression of magnetic or der with oxygen content, several films were subjected to post-growth anneali ng in oxygen ambient (1 atm). Films were annealed at varied temperatures of 600C, 750C, and 900C for an annealing time of 1 hour. This annealing treatment induced a decrease in lattice spacing and a decrease in resistance. Figure 4-3 shows the X-ray diffraction scans for a BaFeO3 film grown at 700C, 100 mTorr O2 before and after a 900C anneal in oxygen. Before annealing, th e peak at 44.21, corre sponding to a cubic dspacing of 4.10 could be deconvolved into two closely spaced peaks. After annealing at 900C, films were measured to be less resistiv e. The resistivity wa s on the order of 100-200 ohm-cm. In addition, the color of the films ch anged from the translucent orange-tan to a reflective brown-black. After annealing, the thin film samples were characterized using x-ray diffraction to quantify changes in structure. Afte r 1-hour annealing of the sample, the peak shifts to 45.13, which corresponds to a smaller lattice spacing of 4.009 Four-circle XRD was used to characterize the in-plane lattice spacing, determined to be 4.0134 The in-plane XRD rocking curve full-width half-maxim um was on the order of 4. Magnetic Properties of Annealed BaFeO3-x Epitaxial Films Significant changes in magnetic properties were observed with annealing. SQUID magnetometry measurements for the as-deposited and annealed films were performed at 10 K, 100K, and 300K. The magnetization behavior for th e as-deposited films c onsidered in Figure 4-

PAGE 57

57 2 shows that they display weak ferromagnetic be havior. In contrast, annealing the films in oxygen induces a robust ferromagnetic phase with a stable coercive field. The magnetization as a function of applied magnetic field was measured for BaFeO3-x films grown on LaAlO3 and SrTiO3 at 600C, then annealed in oxygen at 900C. These films showed clear hysteresis with a coercive field on the order of 800 Oe. Fieldcooled and zero-field c ooled magnetization was measured as a function of temperature. Figure 4-4 shows the magnetization at 5K as a function of magnetic field for films grown on SrTiO3 subjected to various anneal s; clearly, the magnetization increases with annealing in oxygen. Figure 4-5 shows the magnetiz ation as a function of temperat ure for the same films. In this case, a Curie temperature of ~ 235 K is obs erved. Similar experiments were performed for BaFeO3-x films grown on LaAlO3, although the field cooled/zero fi eld cooled splitting was on the order of 100 K for the film on LaAlO3. The realization of ferromagnetism in fully oxidized BaFeO3 and not in the oxygen-deficient material is consistent with the Kanamori-Goodenough rule for superexchange coupling involvi ng transition metal ions, in which the d5-d5 Fe+3O Fe+3 superexchange interaction shoul d be antiferromagnetic, and the d4-d4 Fe+4OFe+4 superexchange interaction ferromagnetic.10,24,25 The difference in magnetic behavior for BaFeO3-x thin films grown on LaAlO3 and SrTiO3 may reflect a strain e ffect due to the differing substrate/film lattice spacing. Chapter 5 looks investig ates this issue. Figure 4-6 shows plots of saturation (Ms), remnance (Mr), Tc, and the peak in the zero field cooled data for the BaFeO3-x films grown on SrTiO3. Clearly, the annealing of BaFeO3-x films results in a ferromagnetic phase with a relatively high value of Tc. While these results are inconsistent with earlier reports of ferromagnetic behavior at 300 K for insulating BaFeO3-x films, they do show that robust

PAGE 58

58 ferromagnetism is observed for highly oxidized pseudo-cubic BaFeO3 at relatively high temperatures. Conclusion In conclusion, BaFeO3-x thin films were grown on single cr ystal perovskite substrates. The as-deposited films had pseudo-cubic perovskite structure, and they were nonconductive and nonferromagnetic, which differed from that of the bulk BaFeO3. Two phases of BaFeO3 were identifiedan oxygen deficient a nd oxygen rich phase, where forma tion is dependent on growth temperature and pressure. With increase in temperature, both phases are apparent. After annealing in oxygen ambient atmosphere, the o xygen deficient phase disappears, and the oxygen rich phase remains. There is a large differen ce in c-axis lattice parameter when comparing asdeposited films to the annealed films. Lastl y, the annealed films showed ferromagnetism and were in some cases conductive. The resi stivity was on the orde r of 100-200 ohm-cm.

PAGE 59

59 20304050607080 10210310410510610710810910101011101210131014 850oC 800oC 750oC 700oCIntensity (a.u.)2-theta (deg) 650oCBFO (100) LAO (012) BFO (200) LAO (024) BFO (300) LAO (312) Figure 4-1. X-ray Diffraction data of as-deposited BaFeO3-x epitaxial films with increasing temperature

PAGE 60

60 -1500-1000-500050010001500 -2.0x10-5-1.5x10-5-1.0x10-5-5.0x10-60.0 5.0x10-61.0x10-51.5x10-5 10K 100K 300KM (emu)Applied Magnetic Field (Oe) Figure 4-2. Magnetization versus applied magnetic field for as-deposited BaFeO3-x grown at 750C and 10 mTorr Fi1

PAGE 61

61 20304050607080 102103104105106107108109 LAO (312) LAO (024) LAO (312) LAO (024) LAO (012) LAO (012) BFO (100) BFO (200) BFO (300) BFO (200) BFO (300)AS-DEPOSITED Intensity (a.u.)2-theta (deg) ANNEALEDBFO (100) Figure 4-3. X-ray diffraction scans for BaFeO3-x films on LaAlO3, both as deposited and after annealing in 1 atm oxygen for 1 hr at 900C

PAGE 62

62 -60000-40000-200000200004000060000 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 As grown anneal 600C anneal 750C anneal 900CMFe (B)H (Oe) Grown at 600C Measured at 5K Figure 4-4. Magnetization versus applied magnetic field for BaFeO3 film grown on SrTiO3 at 600C and 100 mTorr plus annealing in 1 atm oxyge n for 1 hr at various temperatures 050100150200250300350 -5.0x10-30.0 5.0x10-31.0x10-21.5x10-22.0x10-22.5x10-23.0x10-23.5x10-2 MFe(B)T(K) As grown annealed 600C annealed 750C annealed 900C Grown at 6000 C 250 Oe Figure 4-5. Field-cooled and zer o field-cooled magnetization as function of temperature for the BaFeO3-x film after annealing in 1 atm oxygen fo r 1 hr at 900C. The data are taken in a perpendicular field of 30 kOe

PAGE 63

63 Figure 4-6. Magnetization results for BaFeO3 films on SrTiO3. Shows remanence magnetization, saturation magnetization, zero-field behavior and TC as a function of annealing temperature. 02004006008001000 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0 50 100 150 200 250 Peak in ZFC / Tc (K) MFe(B)Annealing temperature Ms (Grown at 600 C) Mr (Grown at 600 C) peak in ZFC Tc

PAGE 64

64 CHAPTER 5 STRUCTURAL CHALLENGES WITH BaFeO3-X EPITAXY Introduction In the previo us chapter, the structural and magnetic properties of BaFeO3-x were investigated and discussed.100 There are many reports on the i nvestigation of oxygen content and various annealing processes in th e synthesis of manganite perovs kite oxides via pulsed laser deposition.101-104 These anneal treatments of perovsk ite compounds containing Mn as B cations include in situ anneal after growth, ex situ anneal in oxygen ambient environments, and growth in various background gases and mixtures to achieve fully stoichiometric compounds.78 However, the same cannot be said about inve stigation of oxygen c ontent in perovskite compounds containing Fe as B cations. Little research has been done on the effect of oxygen content in ferrite perovskite films.90, 105, 106 The growth and characterization of epitaxial BaFeO3-x thin films grown via pulsed laser deposition is revisited in an at tempt to investigate possible stab ilization of its oxygen rich phase as grown, as opposed to growth a nd annealing of these films. In addition, efforts to examine the material challenges that arise af ter post-annealing, for example: thickness issues, films cracking, and attempted forced stabilization via artificial heterostructures, are illustrated. Also, other research group claims concerning the direc tional dependence of magnetic moments will be addressed.20 Experimentation BaFeO3-x thin films were grown on DyScO3pc (3.94), SrTiO3 (3.905 ), LaAlO3pc (3.789 ), and KTaO3 (3.989 ) single crystal substrates vi a pulsed laser deposition. Laser energy densities on the order of 1.4-1.6 J/cm2 and 1.67 J/cm2 were utilized. Though these fluences are similar, the laser output volta ges (in mV) were 580-650mV (high laser energy) and 300mV (low

PAGE 65

65 laser energy). The ablation targets were 1 inch diameter BaFeO3-x ceramics. The thin films were grown at substrate temperatures between 700-750 C and an oxygen pressure ranging from 10500mTorr. After growth, the samples were cooled at 10 /min (and in some cases 20/min) in the oxygen or nitrous oxide partial pressure used during growth and in some cases in situ annealed in at a higher oxygen pressure for 1 hour and then cooled in the in situ annealing pressure. Most films were ex situ annealed in a tube furnace with oxyge n flowing, at anneal temperature at 900C. Also BaFeO3-x/SrTiO3 artificial superlattice films of varying periodici ties were grown, and in some cases ex situ annealed in oxygen ambient atmosphere. The crystal structure and orientation of th e deposited films were examined by X-ray diffraction (XRD). To examine effects of annealing on thickness, images of the sample surface were taken using an OptiPhot Li ght Optical Microscope and record ed as bitmap files. Magnetic characterization was performed on a SQUID magnetometer. Details of substrate and sample preparation are discussed in Chapter 3. Results and Discussion BaFeO3 Epitaxial Films Grow n at Low Laser Energy BaFeO3-x epitaxial films grown on DyScO3, SrTiO3, LaAlO3 single crystal substrates were grown at 700C with varied growth pressures (Po2) and cooled at 20/min in the various growth pressures. It is apparent that when grown at lower laser density, the film s are of high crystalline quality. Shown in Figure 5-1 are x-ray diffraction data, particularly two-theta scans of BaFeO3-x epitaxial films grown on SrTiO3 at the same low laser energy at various pressures. Rocking curves with FHWM values are demonstrated in Figure 5-2. The crystalline quality of the films increases with decrease in pressure, as evidenced by the rocking curve widths. To have the films exhibit ferromagnetic behavior, they are annealed, as stated in the previous chapter. When these films grown at low laser fluence are annealed, they decompose (or degrade) into Fe2O3 and BaO.

PAGE 66

66 Because of this decomposition, further investigations such as magnetization measurements are unnecessary. A brief explanation for this phenomenon will be elaborated on late r in this chapter. BaFeO3 Epitaxial Films Grown at High Laser Energy BaFeO3-x epitaxial films were grown at high laser energy on SrTiO3, LaAlO3 and KTaO3 single crystal substrates. The growth temperature and pressure was at 700C at 100mTorr and cooled at 10/min in the growth pressure. Most films were ex situ annealed at 900C in oxygen ambient environment, and in some cases, films were post in situ annealed cooled in the annealed temperature. These films however, when ann ealed, do not decompose. They still have rather larger rocking curve widths of one degree on averag e. Also, another phenomenon was discovered. After the films were annealed, when examined under a microscope, the fi lms showed directional hatches or cracks. This phenomenon is not si milar to delamination, which occurs when the film literally begins to peel off of the surface of the substrate because of excessive thickness. Shown in Figure 5-3 are optical light microscope images of the surfaces of as-grown and postannealed films, at two thicknesses. The estimated thicknesses (in nm) were 300, 502, 1004, 1580, and 2610. At the second lowest thickness s hown on Figure 5-3B, there are cross marks that occur in a cross diagonal fashion, making the film surface look textured. There is cracking of the film in any place, and so, by visual observati on, this film would look normally flat. However, there is cracking in sma ll areas which start to occur at 502nm in film thickness, and with increasing thickness, there ar e increasing areas of cr acking. At the second greatest thickness, shown in Figure 5-3D, the film has a texture wh ich shows a glitter effect to the visible eye. All of this is mainly attributed to a large change in latti ce parameter upon annealing, and the film cracks on the substr ate due to massive compressive strain.

PAGE 67

67 Effects of film thickness on magnetic properties of BaFeO3 It is important to note that BaFeO3-x epitaxial films grown (at University of Florida) in Chapter 4 were difficult to reproduce at Oak Ridge National Laboratory. Magnetization values shown here in Chapter 5 are never as high as those values in Chapte r 4. Figure 5-4 shows varying thickness (in number of pulses) of as-deposited films synthesized at laser energies of 580-625mV. Shown are three films with large differences in the number of laser pulses which demonstrate the definite effect that thic kness has on the crystalline quality of BaFeO3-x epitaxial films. With this effect noted, the magnetic properties of films at varying thickness were explored. Figure 5-5 indicates the magnetizatio n of five films varying from 1250 to 20,000 (or ) laser pulses which correlates to roughly 300 nm to 3,000 nm (or 10) in thickness. Note that these films were all anneal ed at 900C in oxygen ambient. Surprisingly, with decrease in film thickness there is decrease in magnetization. As mentioned in the previous section, the higher crystalline films (grown at lower laser fl uence) were not convertible. In this case, convertible refers to films that have optimal conditions to easily become magnetic after ex situ annealing. Consistently, the le ss thick films which are also the films with higher crystalline quality, also are more difficult to become convertible films, based on the weaker magnetic moment (and aside from being thinner films) Figure 5-6 shows the trend of magnetization versus thickness of the films. Note that there is a threshold or minimum thickness for films to become convertible, in this case, greater than 50 2nm but less than or equal to 1004nm, which is a small window for thickness variation. In addition, referring back to Figure 5-4, the film grown at 205 nm has a rocking curve width of 0.351, sign aling high crystalline quality, and the film grown at 1004nm has a rocking curve width of 0.725, evidence of low crystalline quality, because standard high crystalline films will usually have rocking curv e widths of less than 0.5.

PAGE 68

68 Effects of oxygen and nitrous oxide partial p ressure on structure of BaFeO3 There were attempts at forcing BaFeO3-x to fully oxidize when as grown, as opposed to undergoing ex situ anneal treatment. Films grown in 100mTorr Po2 were compared to films grown in 100mTorr PN2O. The following is a discussion on the effects of different partial pressure and substrate choice on structure. In the two-theta scans, there are usually tw o peaks known as peak A (oxygen deficient) and peak B (oxygen rich), see for example Figure 4-1. The ratio of the intens ities of the two peaks depends on the substrate they are grown on (or th e lattice mismatch). With increasing lattice mismatch the ratio of the intensities of the oxyge n-deficient to oxygen-rich decrease. In Figure 5-6, BaFeO3-x films were grown at the same conditions, but on three different substrates. On KTaO3 there is only one peak visible, the oxygen deficient peak. On SrTiO3 both peaks are visible, with the oxygen-rich peak having a much higher inte nsity than the oxygen deficient peak. Many films were grown on SrTiO3, but only a few (two-three) showed opposite ratio intensities. On LaAlO3, the ratios for the most part would be equal or the oxygen-deficient peak would be slightly lower intensity th an that of the oxygen-rich peak. Reciprocal space maps in Figures 5-7 5-8 show interesting phenomena as well. Figure 57 is of two reciprocal space maps of as-grown BaFeO3-x on STO, grown in 100 mTorr Po2 and in 100mTorr PN2O. Nitrous oxide (N2O) promotes high oxidizing conditions. Shown in Figure 58, is BaFeO3-x grown on three different substrates in 100mTorr PN2O. With the largest lattice mismatch (LAO), what looks like unusual struct uring occurs, but decr eases with decreasing lattice mismatch. These structur es all lie on the same c-axis d-spacing, while having different a (=b) axis d-spacing, which is puzzling. On STO, the unusual structuring is not strongly present. On KTO, only a single peak remains (signify ing the oxygen deficient phase domination). One possible explanation for this could be a chemi cal separation of some kind and possible phase

PAGE 69

69 different orientations arising. Table 5-1 shows the lattice mismat ch between the two phases of BaFeO3-x and the substrates used and other substrates for further comparison. Note the huge lattice mismatch of BaFeO3-x with that of LAO, and STO, while with KTO, the lattice mismatch is almost 7% less than that of LAO. This can support the notion that with greater substrate mismatch, BaFeO3-x is less crystalline in qua lity, yet more convertible to a magnetic material. Lastly, it is widely known that to obtain higher quality crystalline films, lattice mismatch should be kept to below 5%, of which the mismatch between BaFeO3-x with LAO and STO are both 5%. Magnetization in Different Directions In a paper by Taketani et al,20 the authors measured the magnetization of films that were grown via pulsed laser deposition and in differe nt planar directions. They measured the magnetization perpendicular to the plane (100), parallel to the pl ane (001) and in-plane (110). They concluded that the magnetization in each di rection was different and therefore the sample exhibits conventional antiferromagnetic behavior with magnetic spins along the (100) crystal axis.20 Here, the magnetic properties of annealed samples of BaFeO3-x on SrTiO3 and KTaO3 single crystal substrates were inve stigated in the (100) and (001) directions. Shown in Figure 5-9 is the magnetization versus field of samples measured with crysta l orientation pe rpendicular to the field and parallel to the fiel d. Concluded here is that in these films there is no observed change in the strength or the coercivity of the magnetizati on, hence it does not depend on direction, as the previously mentioned authors stated.20 In addition to this, the papers the authors cited107, 108 were of samples having 3-5 B per Fe, a few orders of magnitude higher than that of the authors samples. Moreover, both of those cited papers showed evidence of helical ordering

PAGE 70

70 from neutron diffraction studies, and these samp les were limited to room-temperature SQUID magnetometry. Lastly, recent neutrons diffr action experiments were executed to further confirm that antiferromagnetic ordering is not present in BaFeO3-x. Neutron data showed small peaks, with no temperature dependence, possibly structural peaks. However, no strong antiferromagnetic structure peaks were observed, which shows that BaFeO3-x is not antiferromagnetic, but is rather, a large and disordered material, structurally and magnetically. Artificial Superlattices of SrTiO3/BaFeO3 To further elucidate on this cracking phenomena, SrTiO3/BaFeO3 artificial superlattices with periodicities of 500nm (bi-layer), 10nm, and 4nm were synthesized. This experiment was conducted to determine if mini mizing the thickness of BaFeO3 would ease the phenomenon of cracking after ex-situ anneal. The artificial s uperlattices were grown via alternation of a continuous two-target rota tion system, using BaFeO3 and SrTiO3 ceramic targets. Shown in Figure 5-10 is a 2t-w scan of an attempted [STO5nm/BFO5nm] artificial superl attice that is 500nm thick on LaAlO3 and SrTiO3 substrates. From the defined satell ite peaks, the periodicity of these heterostructures is = 10 nm. In the inset is a reci procal space map of the corresponding superlattice film on LaAlO3. The satellite peaks are well-defin ed, and interestingly, they cluster around the relative d-spacings where BaFeO3 and SrTiO3 would each be located had they been each synthesized separately onto the substrate. Figure 5-11 shows x-ray diffraction scans of attempted 4nm peri odic artificial superlattices: as-grown, and then post-annealed. The measured periodicity of the as-grown heterostructure was 2.4nm. After annealing, the periodicities were again calculated from 2tw scans to be 2.7nm. The periodicities had increased meaning the overall thickness of the

PAGE 71

71 heterostructure increased. In a normal single epitaxial BaFeO3-x film, annealing changes decreases the out-of-plane lattice parameter, while in most cases the in-plane lattice parameter remains the same. So, in a superlattice, with periodicity n, the individual material thickness layers are b and s (b = BaFeO3-x and s = SrTiO3), making b + s = n. So after annealing this heterostructure, the periodicity must now be: (b d ) + s = n, where d is the change difference in the out-of-plane lattice parameter of BaFeO3-x. However, this is not what is seen according to the two-theta scans. Reason for this inconsistenc y must be attributed to the placement of the substrates. Illustrated in Figure 5-12 is a post-depositi on schematic of the substrate holder with substrates still adhered, and the footprint of th e deposition. Note that depending on where the substrates are positioned, that there is a deviance in the accuracy of the attempted periodicity of 4nm artificial superlattices. Note also that subs trate size also plays a part in the accuracy of periodicity, because there most likely gradient diffe rences in the thickness of the layers as one goes from the center of the depositio n area out toward the edges. Moreover, annealed artificial superlattices of all the men tioned periodicities, down to = 4 nm, exhibit cracking phenomena as shown in Figure 5-13. This pr esents further evidence that the cracking phenomena of BaFeO3-x does not depend on thickness of BaFeO3-x. Even with thin alternating layers of BaFeO3-x and SrTiO3, it is mainly the large cha nge in lattice constant upon annealing that causes cracking of BaFeO3-x. Conclusion Further investigation of BaFeO3-x epitaxial films demonstrated complex domain structure, as seen in reciproc al space maps. After ex situ anneal in oxygen, these co mplex structures clean up into a single pseudocubic stru cture that is magnetic. BaFeO3 grown under low laser fluence

PAGE 72

72 is not convertible via anneal to a magnetic material. Its phase composition decomposes into BaO and Fe2O3. With thickness, the crystall ine quality of as-grown BaFeO3-x decreases, but this decrease makes it a more compatible material to anneal and become magnetic. Artificial superlattices of SrTiO3/BaFeO3 were synthesized to attempt to decrease cracking of annealed BaFeO3, which occurs due to a large change in la ttice constant when converted to a magnetic phase. Periodicities down to 4nm continued to re sult in partial cracking of the multilayers. The magnetization of ex situ annealed BaFeO3-x epitaxial films were ex amined as a function of direction, and was found not to have change in ma gnetization with direction of field with respect to crystalline plane, desp ite other research claims.20 BaFeO3-x is speculated to be highly disordered, structurally and magnetically.

PAGE 73

73 424344454647484950 SrTiO3As-grown BaFeO3-x on STO 700oC; 12,000 pulses 1 mTorr 100 mTorr 300 mTorr Intensity (arb. units)Angle 2(degrees) Figure 5-1. Two-Theta-Omega scans of as-grown BaFeO3-x thin films at various pressures.

PAGE 74

74 20.020.521.021.522.022.523.023.524.0 0.507o0.42o1.273o 1 mTorr 100 mTorr 300 mTorr Intensity (arb. units)Angle (degrees) Figure 5-2. Omega Rocking Curves show FWHM of (002) peaks of the films at various pressures.

PAGE 75

75 Figure 5-3. Optical Light Microscope images of surfaces of as-deposited and annealed BaFeO3 films. Shown are: A) as-grown film at 1580 nm, B) annealed film at 1580 nm, C) asgrown film at 502 nm, D) annealed film at 502 nm

PAGE 76

76 10.210.510.811.111.411.7 0.351o0.725o 205 nm 1004 nm 1580 nm Intensity (arb. units)Angle (degrees) 1.136o Figure 5-4. Omega Rocking Curv es of as-deposited BaFeO3 epitaxial films at different thicknesses.

PAGE 77

77 -60000-40000-200000200004000060000 -0.35 -0.30 -0.25 -0.20 -0.15 -0.10 -0.05 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 10K M(B/Feion)Field (Oe) 2610 nm 1580 nm 1004 nm 500 nm 300 nmFigure 5-5. Magnetization versus field data of annealed BaFeO3-x epitaxial films at varying thicknesses.

PAGE 78

78 050010001500200025003000 0.025 0.050 0.075 0.100 0.125 0.150 0.175 Remnant Magnetization, Mr (B/Fe)Film Thickness (nm) Figure 5-6. Plot of remnant magnetization, Mr versus film thickness (in nm).

PAGE 79

79 A B Figure 5-7. Reciprocal Space Maps of a) BaFeO3-x on STO at 100mTorr Po2, and b) BaFeO3-x on STO at 100mTorr PN2O

PAGE 80

80 A B Figure 5-8. As-deposited BaFeO3-x at 100mTorr PN2O namely on A) LAO, B) STO, and C) KTO

PAGE 81

81 C Figure 5-8. Continued

PAGE 82

82 Table 5-1. Comparison of lattice constant mismatch with the two phases of BaFeO3-x Substrate/Buffer Film BaFeO3-x % Mismatch (O2 Def: a= 4.1 ) BaFeO3-x % Mismatch (O2 Rich: a = 4.012 ) LaAlO3 (3.789 ) 8.2% 5.89% SrTiO3 (3.905 ) 4.99% 2.74% DyScO3 (3.94 ) 4.06% 1.83% KTaO3 (3.989 ) 2.78% 0.57% SmScO3 (3.990 ) 2.76% 0.55% BaZrO3 (4.186 ) 2.05% 4.15%

PAGE 83

83 -60000-40000-200000200004000060000 -0.315 -0.210 -0.105 0.000 0.105 0.210 0.315 T= 5K 10k shots 001 10k shots 100T= 10K Magnetization (b/Fe ion)Field (Oe) Figure 5-9. Magnetization versus Magnetic field data of BaFeO3-x measured parallel and perpendicular to the crystal axis

PAGE 84

84 Figure 5-10. X-ray diffraction of [SrTiO3/BaFeO3] artificial superlattices on LAO and STO substrates. Inset is reciprocal space map of superlattice on LAO.

PAGE 85

85 Figure 5-11. X-ray diffraction data of as-grown a nd annealed [STO/BFO] arti ficial superlattices. Attempted periodicity is 4nm.

PAGE 86

86 Figure 5-12. Schematic of substrate holder afte r film deposition with s ubstrates attached. Substrates are labeled with corresponding periodicities () of the [STO/BFO] superlattice growth.

PAGE 87

87 A B Figure 5-13. Optical microscope images of ar tificial superlattices, namely A) as-grown [STO/BFO] and B) a nnealed [STO/BFO].

PAGE 88

88 CHAPTER 6 SYNTHESIS AND PROPERTIES OF (Bi,Ba)FeO3 SOLID SOLUTION THIN FILMS Introduction There is g reat interest in multiferroics: materials that possess magnetic and ferroelectric properties in the same phase. These materials have potential for applications due to the possibility of important physical properties leading to a new generation of memory devices.5 As an interesting and widely researched multiferroic, BiFeO3 exhibits a large spontaneous polarization below 1100 K and is antiferromagnetic be low 640 K. Its compatibility toward the existence of these two ferroic pr operties is due to its unique Bi lone pair mechanism, which doesnt interfere with the antiferromagnetically aligned Fe ions. The unique properties of BiFeO3 consistently occur in bulk, an d as epitaxial films grown at all thicknesses at optimal conditions.36, 72 BiFeO3, like most other single phase existing multiferroics, has a setback in which it does not have a net magnetic momentdue to its helical magnetic ordering, which if existed, would make BiFeO3 more attractive for device applica tions. Research groups have in recent years begun studying the effect of Aand B-site substitution and solid solutions in the BiFeO3 system in an attempt to resolve the issue of mostly zero net magnetization.32, 53, 75, 109 In particular, their findings show th at A-site substitution of diama gnetic large ionic radius ions show most promise in bringing an increase in magnetization values.62-64, 67, 70, 71 Discussed previously in Chapters 4 and 5 is the study of the magnetic properties of epitaxial films of BaFeO3-x,100 an oxide that exists as a hexa gonal structure in bulk, but can be stabilized as a perovskite oxide when grown as an epitaxial film. BaFeO3-x can have ferromagnetic properties depending on growth and annealing conditions.20, 22, 92, 100 The properties of BaFeO3-x show weakly ferromagnetic behavior and, in addition, the material is highly insulating. Thus BaFeO3, mixed with BiFeO3, can show interesting properties, possibly

PAGE 89

89 without compromising its ferroelect ric properties. As discussed in Chapter 2, there are a few groups that have studied the properties of A-site cations, specifically a few alloy compositions of substitution of Ba in BiFeO3. However, all of these studies have been carried out only in bulk sample experiments.65, 66, 68, 69 Given the magnetic ordering in both BaFeO3 and BiFeO3, this work examines structural properties of the creation of (BixBa1-x)FeO3 solid solutions epitaxial films via pulsed laser deposition on single crystal substrates. Experimentation Synthesis of (BixBa1-x)FeO3-x Films via Continuous Alternating Target Rotation There were two ways the alloys of (Bi,Ba)FeO3-x,were synthesized. In the first method, the growth modes and conditions were found fo r each of the end members. Because BiFeO3 is a well established multiferroic material which has a strong ferroelectric polarization above room temperature, it is logical to star t at its optimal growth condition s and tune the growth of BaFeO3-x around the conditions for BiFeO3 growth. The conditions for BiFeO3 are discussed elsewhere.36 After reaching compatible conditions, the soli d solution films were grown by pulsed laser deposition via two-target continuous rotation, previously described in Chapter 3, on DyScO3 (3.94 ) SrTiO3 (3.905 ), and LaAlO3 (3.789 ) single crystal substrates. Synthesis of (BixBa1-x)FeO3-x Films via Solid Solution Targets The second experimental method to synthesize alloys of (Bi,Ba)FeO3-x, was via solidsolution targets. Instead of mixing powders of BaFeO3-x and BiFeO3 together, the primary method for making BaFeO3-x targets was used (the calcination of loose BaCO3 mixed with Fe2O3) and then Bi was added to the powder depending on the alloy desired, and the powder mixture was pressed into a pellet and sintered. Three alloy targets were made: Bi0.1Ba0.9FeO3-x, Bi0.5Ba0.5FeO3-x, and Bi0.9Ba0.1FeO3-x. After completion, the three ceramic targets, each with different compositions, were examined with X-ra y diffraction. Figure 6-1 illustrates structural

PAGE 90

90 information of the targets, noting that they are acceptable with respect to the reference positions of the end members and BaFeO2.5. In addition, the targets Bi to Ba composition ratios was examined using EDS analysis and was within a marginal error percentage. Lastly, no buffer layer was used in growing the films of this second method. The only change in growth conditions between the two-target method and the solid-solution target me thod was a decrease in laser fluence to 1.27 J/cm2 in the solid-solution target met hod. The solid solution films were grown by pulsed laser deposition at low laser energy on LaAlO3 (3.789 ), NdGaO3 (3.85 ), (La0.29,Sr0.71)(Al0.65,Ta0.35)O3 (LSAT) (3.86 ), and SrTiO3 (3.905 ) single crystal substrates. The crystal structure and orientation of th e deposited films were examined by X-ray diffraction (XRD). Magnetic characterization was performe d on a superconducting quantum interference device (SQUID) magnetometer. Transmission Electron Microscopy (TEM) was performed on some of the samp les by collaborating authors. Results and Discussion (BixBa1-x)FeO3-x Films via Continuous Alt ernating Target Rotation The midpoint of the solid solution, Bi0.5Ba0.5FeO3-x, was created as a starting point. Epitaxial Bi0.5Ba0.5FeO3-x films on (001)pc DyScO3 and (001)pc LaAlO3 were fabricated using the continuous two-target rotation method. Though not shown, the films grow in on the pseudo cubic substrates, and have no extr a diffraction peaks other than th e (00L), indicating single phase epitaxy. Figure 6-2 shows varied thic knesses of the 50/50 alloy grown on DyScO3 single crystal substrate at a temperature of 700C and pressure of 50m Torr. Ther e is a shift to the right in dspacing with increasing thickness. This may indicat e that there could be strain occurring in the films at a thickness of 80nm or less. To mimic the surface and lattice constant of DyScO3, a relatively thick layer (100nm) of SmScO3 on LaAlO3 was grown to serve as the surf ace of the smaller lattice constant

PAGE 91

91 substrates. This buffer reduces the lattice mi smatch between the solid solution alloys and the LaAlO3 substrate. Figure 6-3 show s structural data for the 50/ 50 alloy composition with varied thicknesses, this time using the SmScO3 buffer on LaAlO3 substrates. At all three thicknesses there are no extra diffraction peaks other than (00L) that show up that would indicate secondary phases. Again, these epitaxial films may be stra ined at 80 nm due to the similarly comparable lattice parameter of the SmScO3 buffer, but this is can only be quantitatively determined by reciprocal space maps. Figure 6-4 contains rocking curve measurem ents on the three films. At all three thicknesses, the rocking curve wi dths are below 0.3 degrees, indica ting high crystalline quality. Although it is not completely evid ent in the two-theta scan and omega rocking curves, according to reciprocal space map (RSM) measurements not shown, Bi0.5Ba0.5FeO3 films are strained to the SmScO3 buffer with aand c-axis parameters of 3.993 and 4.063 at a thickness of 80nm. At thicknesses of 160nm, the films are relaxed with aand c-axis lattice parameters of 4.00 and 4.07 At 240nm, aand c-axis parameters ar e 4.00 and 4.06 virtually no change in lattice constant (between the film grown at 160nm and 240nm). This is ev idence of complete relaxation of the film. Alloys of compositions were successfully fabricated via th e two-target rotation method on LaAlO3 substrates with relaxed SmScO3 buffer film. Figure 6-5 shows the x-ray diffraction results for the solid solution films at 0%, 10% 20%, 30%, 40%, 50%, 60%, 70%, and 80% Bi content. The diffraction results in Figure 6-5 show a continuous shift to the right in d-spacing, with the alloy film peaks (furthes t left) moving closer to the SmScO3 buffer (film) peak. All of the alloys showed no extra peaks other th an (00L) indicating si ngle phase epitaxy.

PAGE 92

92 Using four-circle X-ray diffracti on, reciprocal space maps of the epitaxial films, buffer, and substrate at 20%, 50%, and 70% Bi c ontent are shown in Figure 6-6. The SmScO3 buffer was grown thick enough to be relaxed in most cases (more than 80nm in thickness). With increase in Bi content, the film peak begins to line up with that of the SmScO3 buffer, signaling a decrease in aand clattice parameter. A clear difference in Figure 6-6A and Figure 6-6C is seen where the increase in Bi, th e smaller cation, has re duced both the a-axis an c-axis lattice constant. Figure 6-7 shows the in-plane and out-of-pla ne lattice parameters extracted from the reciprocal space maps. There is a cons istent trend in lattice constant until at x 0.5 when the solid solution epitaxial films start to become cohe rent with that of the buffer. After x > 0.4, the in-plane lattice parameters hover around that of the SmScO3 lattice parameter while the out-ofplane (or c-axis) lattice parameters continue to shrink, indicating a shrinking in tetragonality to a more cubic structure. From x 0 to x 0.4, the a-axis lattice para meter is actually shrinking almost the same amount that the c-axis para meter is shrinking, indicating strong tetragonal behavior in these alloys. This is also shown in Figure 6-8, wh ere the c/a ratios show that the structure varies from highly tetra gonal to a basic cubic structure, and a transition is observed at x 0.5. Note that BaFeO3 is found to be having usually te tragonal structure, while BiFeO3 varies from rhombohedral to slightly tetragonal depending on the film thickness. (BixBa1-x)FeO3 Films via Solid Solution Targets Referring back to Figure 6-1, the ceramic target s were characterized via x-ray diffraction. In the 2-theta scan, structural data of three targets, Bi0.9Ba0.1FeO3-x, Bi0.5Ba0.5FeO3-x, and Bi0.1Ba0.9FeO3-x, are shown. All peaks (except one at 2theta = 28 degrees), were able to be identified. This was carefully executed by using the JCPDS International tables of structural data for BaFeO3 and BiFeO3 and matching the (hkl) with the peak intensities shown on the

PAGE 93

93 targets structural patterns in the two-theta scan. The tre nd in increasing Bi content and decreasing Ba content is evident via the shift of the peaks in each diffraction pattern. In the insert, this trend is fully illustra ted with the labeling of where BaFeO2+x, BaFeO3-x, and BiFeO3 (111) peak would be, respectively. Thin films of Bi0.9Ba0.1FeO3-x were successfully synthesized via pulsed laser ablation from a Bi0.9Ba0.1FeO3 solid solution ceramic target. Ta king a closer look at the alloy, Bi0.9Ba0.1FeO3-x, was grown on five different substrates in or der of increasing lattice parameter: LaAlO3 (3.789 ), NdGaO3 (3.85 ), (La0.29,Sr0.71)(Al0.65,Ta0.35)O3 (LSAT) (3.86 ), SrTiO3 (3.905 ), and KTaO3 (3.99). The films are epitaxial, as illustrate d in Figure 6-9. There is a large change in d-spacing depending on the substrate the film was gr own on. In Table 6-1 is the measured c-axis d-spacings of the film and substrate peaks, taken from the two-theta scan. It clearly shows the extreme change c-axis d-spacing with choice of substrate. The measured film c-axis parameters of the four films range from 4.6007 and d ecrease to 4.161, a drastic difference. Upon closer examination of the two-theta sc ans, the films on LSAT and on NGO show two peaks, similarly crossing over each other. It ap pears as a possible transition; and it could be structural and/or chemical. Fi gure 6-10 displays the Omega rocking curves of the thin films. The full-width-half-maximum values are labeled beside each peak. The rocking curves show evidence of the distribution of crystalline domains. Equally important, th e rocking curve widths for all four of the epitaxial films fall below 0.22 degrees, indicating highly crystalline films. The largest rocking curve, which is on L AO at a FWHM of 0.215, is surprising since there should be a rather larg e lattice mismatch between Bi0.9Ba0.1FeO3 and the LAO substrate. However, because Bi0.9Ba0.1FeO3 was first synthesized in this work, no prior structural data exists on it, so it is impossible to find know be forehand its actual c-axis lattice parameter. A

PAGE 94

94 speculation of the potential mismatch between LAO and Bi0.9Ba0.1FeO3 can be estimated from one of its parent compounds, BiFeO3. BiFeO3s c-axis parameter (relaxed) is approximately 3.962. So taking this value, and doing calculations, the lattice mi smatch with LAO substrate is about 4.4%, which is pretty high. Assuming Bi0.9Ba0.1FeO3 will have a larger lattice parameter than BiFeO3 due to the presence of large Ba cations, the lattice mismatch will definitely be higher. Also note that the FWHM of Bi0.9Ba0.1FeO3 on STO is not much smaller than that on LAO; its rocking curve width is around 0.1. This is a tiny difference in FWHM for such a big difference in film-substrate lattice mismatch (BiFeO3s lattice mismatch with STO is around 1.4%). With further investigation, the in-plane and out-of-pl ane lattice parameters of these films along with the corresponding substrates were ex tracted from reciprocal space maps. From studying the reciprocal space maps in Figure 6-11, all four epitaxial films ar e entirely strained to the substrates. This is evidenced by the position of the film (103) peak being directly under the (103) peak of the substrate. This is interes ting as these films are ap proximately 175nm, which is rather thick for complete strain to be present. Also note the increasingly large change in c-axis lattice parameter of the strained films with decreasing substrate lattice parameter. Table 6-2 demonstrates the complete picture of these excessively strained thick epitaxial films. It is interesting that the unit cell volu me of these samples is increasingly larger with decreasing lattice parameter of th e substrate (also equal to the in-plane lattice parameter). More importantly, the in-plane lattice parameter only matches that of the substrates exact lattice value, and this occurs with all of the substrates. Note that there are two values for the films grown on LSAT and NGO due to the presence of dual peak s in both the two-theta scans and reciprocal space maps. Also listed in Table 6-2 is a second (hkl) reflection for Bi0.9Ba0.1FeO3 on LAO. The

PAGE 95

95 LAO (113) peak shows comparable values to th e LAO (103). Furthermore, information for two different samples of Bi0.9Ba0.1FeO3 grown on STO is shown in the table in case of any doubts of reproducibility, because admittedly, these samples prove to be highly unusual in the scope of epitaxial film research. It should be mentioned that while KTO is labele d on this table, there is no structural data for Bi0.9Ba0.1FeO3 on KTO. When this film was grown on th is substrate, the two-theta scans only showed that the KTO substrate peaks were visible, however no film peaks were visible. Twotheta scans measured from 0.2 out to 110 also showed no presence of Bi0.9Ba0.1FeO3 peaks (or any others). This can mean two things. The first is that the fi lm did not grow (or have epitaxy) on KTO. The second is that the film grew, but is completely embedded in the substrate. This is possible, knowing that the esti mated lattice parameter of Bi0.9Ba0.1FeO3 is incredibly close to that of KTO (3.99). Reciprocal space maps were done for this sample which didnt show any conclusive evidence that a film p eak was there and/or whether it wa s strained to the substrate. This is the reason for question marks for the KTO sample in Table 6-2. To have a quantitative perspective of the extreme lattice parameters of these completely strained epitaxial films, Figure 6-12 graphs the backed-out lattice parameters (both aand c-axis) of the films. For the Bi0.9Ba0.1FeO3 film on KTO, for illustration purposes, this value for the caxis is estimated and under assumption that the film is embedded in the substrate. In Figure 612, there are two trends and where the trends meet is at the transition at Bi0.9Ba0.1FeO3 films on LSAT and NGO. This trend is also shown in Fi gure 6-13, where the c/a ratios versus substrate values illustrate extreme tetragonality and then reduce down to an almost cubic ratio. Again, for the Bi0.9Ba0.1FeO3 film on KTO, this is estimation. These numbers seem hard to believe to have perovskite structure, so Transmission Elect ron Microscopy (TEM) images of two samples

PAGE 96

96 (Figures 6-14 and 6-15) illustrate the that these ar e indeed real values for real samples. Figure 614 is a TEM image of Bi0.9Ba0.1FeO3 on LSAT where the film is th e light colored area and the darker area is the substrate. At the interface, th e film is coherent with the substrate, relatively clean, showing that there is no major defect occu rring there. In Figure 6-15, the interface of Bi0.9Ba0.1FeO3 with the LAO substrate is also sharp. Looking at the film (lighter area) the film is single phase, and no major defects ar e shown which supports that the Bi0.9Ba0.1FeO3 epitaxial film on LAO is perovskite. The magnetic properties of as-grown films are demonstrated in Figure 6-16. The magnetization versus magnetic field data s how weak ferromagnetic hysteresis in Bohr Magnetons per Fe for consiste ncy. The remnant magnetization Mr of all of the samples is 0.2B per Fe. This magnetic data is comparable to th at of other groups whom measured magnetizations of Ba-doped bulk BiFeO3 samples65, 68 The coercivity, or resistance of the sample to demagnetization, is about 6000 Oe on LAO, 8000 Oe on STO, and about 1 Tesla on LSAT, with a magnetic moment for all three, comparable to annealed BaFeO3-x and obviously greater than that of BiFeO3 epitaxial films.36 It is important to note is that having comp arable magnetic moment only after adding only roughly 10% Ba to pure BiFeO3 is surprising and in teresting. This means that with a small amount of Ba ions, and them being far apart from each other, supports the claim that was made in Chapter 5 that BaFeO3 is a highly structurally and ma gnetically disordered material. Theoretically, highly oxidized BaFeO3-x is calculated to have magne tic moments of a lot more than at 0.2B. Those films are of low crystalline quality, and it makes sense that they also have low magnetic moment. Moreover, the Bi0.9Ba0.1FeO3 epitaxial films are highly crystalline, and so it is less likely they are magnetically disordered as well.

PAGE 97

97 The apparent pinching at 1 Tesla and -1 Tesla is interesting to view. The pinch of the hysteresis loops is at the same magnetic fields in all of the samples. BiFeO3 is known to possess G-type antiferromagnetic helical structuring which is complex.33 In previous works other groups claim that with substitution of certain A-site cations into BiFeO3 and upon measuring small magnetic moments and obtaining very weak magnetic hysteresis, they are, in effect, breaking down the spiral structure that is responsible for the antiferro magnetism present in pure BiFeO3.32, 36, 53, 63-71, 75, 109 One can speculate, that this pinchi ng of the magnetic hysteresis loops is directly related to the disruption of helic al magnetic ordering via introduction of 10% Ba2+ cations into BiFeO3, and could be cited as evidence for such phenomena.23, 33 Conclusion Alloys of a solid-solution between BiFeO3 and BaFeO3-x have been successfully created. The alloy films that were create d via two-target rotation method ha ve been carried all the way to 80% solubility. The alloy films created via solid-solution targets ha ve been successfully fabricated at near both end-memb er-points and at the half-point showing that the solubility is possible over the entire range of the solid-solution. Bi0.9Ba0.1FeO3 epitaxial films are of high crystalline quality with rocking curves widths of less than 0.2, are fully strained, and have highly unusual in-plane and out-of-pl ane lattice parameters, as well as extremely high c/a ratios. TEM shows that the films on LAO are mostly single phase and support evidence for having perovskite structure. Investigation of the magnetic properties reveals that the samples are ferromagnetic (or ferrimagnetic) with a magnetization of 0.2B per Fe. This is a huge increase (more than an order of magnitude) over that of pure BiFeO3. With unfamiliar pinching of the magnetic hysteresis loops, there may be possible breaking down of the spiral magnetic ordering in the largely strained samples.

PAGE 98

98 Figure 6-1. Structural data of three BixBa1-xFeO3 ceramic targets. Points of BiFeO3, BaFeO3-x, and BaFeO2.5+x are labeled for reference.

PAGE 99

99 Figure 6-2. X-ray diffraction data for varied thicknesses of Bi0.5Ba0.5FeO3-x on DyScO3 substrates.

PAGE 100

100 444546474849 101102103104105106107 LaAlO3 Bi 0.5Ba0.5FeO3-x 80 nm 160 nm Intensity (arb. units)Angle 2 (deg) 240 nm SmScO3 (100nm)(700C, 50mTorr, varied thickness) Figure 6-3. X-ray diffracti on (2-theta scan) for Bi0.5Ba0.5FeO3-x alloy films at various thicknesses with SmScO3 buffer. 21.321.621.922.222.522.8 0.2o0.277o0.166oIntensity (arb. units)Angle (deg) 240 nm 160 nm 80 nm Figure 6-4. X-ray diffraction data for Bi0.5Ba0.5FeO3-x alloy films at various thicknesses. Omega rocking curves of thin films show ing strain/relaxation behavior.

PAGE 101

101 4042444648505254 10-11001011021031041051061071081091010101110121013 ** 0% Bi, or 40% 60% 70% 10% Intensity (arb. units)2 (degrees)(BaFeO3-x) 20% 30% 50% 80% SmScO3LaAlO3*from x-ray source **WL1 rad Figure 6-5. X-ray diffraction data for solid solution (BixBa1-x)FeO3 grown by pulsed-laser deposition.

PAGE 102

102 A 0.240.260.28 0.72 0.75 0.78 LAO 103 Bi0.2Ba0.8FO 103 1/d100 (A-1)1/d003(A-1) SSO 103 B 0.24 0.26 0.28 0.72 0.75 0.78 Bi0.5Ba0.5FO 103 SSO 1031/d100 (-1)1/d003 (-1)LAO 103 C 0.240.260.28 0.72 0.75 0.78 1/d100 (A-1)1/d003 (A-1) LAO 103 Bi0.7Ba0.3FO 103 SSO 103 Figure 6-6. Reciprocal sp ace maps for selected (Bix Ba1-x)FeO3 films on LaAlO3 with SmScO3 buffer. A) (Bi0.2 Ba0.8)FeO3 B) (Bi0.5 Ba0.5)FeO3 C) (Bi0.7 Ba0.3)FeO3

PAGE 103

103 0.00.10.20.30.40.50.60.70.80.91.0 3.990 4.005 4.020 4.035 4.050 4.065 4.080 4.095 4.110 4.125 4.140 4.155 4.170 4.185 in-plane out-of-plane ------SmScO3 Buffer Lattice constant (A)Bi content (x) Figure 6-7. In-plane and out-of-plane lattice pa rameters extracted from the reciprocal space maps. 0.00.10.20.30.40.50.60.70.80.91.0 1.000 1.004 1.008 1.012 1.016 1.020 1.024 1.028 c/a RatioBi content (x) Figure 6-8. C/A ratios calculated from lattice constants extracted from the reciprocal space maps.

PAGE 104

104 Figure 6-9. Bi0.9Ba0.1FeO3-x epitaxial films grown on f our different substrates. Table 6-1. Measured c-axis dspacings of substrate and corr esponding films, taken from twotheta scan. Substrate Substrate (c-axis) measured () Film (c-axis) measured () LAO 3.788 4.6007 NGO ~3.86 4.412 LSAT ~3.86 4.245 STO 3.904 4.161 KTO 3.989 3.969 ( ? )

PAGE 105

105 Figure 6-10. Omega rocking curves with measured widths illustrate highly crystalline phases.

PAGE 106

106 0.220.240.260.280.30 0.70 0.72 0.74 0.76 0.78 BBFO (103)1/d0031/d100CC072808STO-A (103)STO (103) A 0.200.220.240.260.280.30 0.68 0.70 0.72 0.74 0.76 0.78 0.80 BBFO (103)1/d0031/d100CC072808LSAT-ALSAT (103) B Figure 6-11. Reciprocal space maps of Bi0.9Ba0.1FeO3-x epitaxial films grown on four different substrates, namely A) STO, B) LSAT, C) LAO, and D) NGO

PAGE 107

107 0.200.220.240.260.280.300.320.34 0.64 0.66 0.68 0.70 0.72 0.74 0.76 0.78 0.80 BBFO (103) CC072108LAO-A (103)1/d0031/d100LAO (103) C 0.220.240.260.280.300.32 0.66 0.68 0.70 0.72 0.74 0.76 0.78 0.80 BBFO (103)1/d0031/d100CC072808NGO-A NGO (103) D Figure 6-11. Continued

PAGE 108

108 Table 6-2. Measured and calcula ted in-plane and out of plane lattice parameters, and unit cell volumes for selected epitaxial films (calcu lated from reciprocal space maps or RSMs). Sample Number & hk l Reflection suba Filma=b Filmc Cell Volfilm 0721-LAO-103 3.782 3.798 4.620 66.652 0721-LAO-113 3.792 3.792 4.622 66.45 Not shown NGO-103-to p ~3.85 3.844 4.226 62.4 NGO-103b ott. 3.844 4.411 65.16 LSAT-103-to p ~3.86 3.861 4.240 63.202 LSAT-103b ott. 3.861 4.377 65.260 0721-STO-103 ~3.903 3.903 4.137 63.023 Not shown 0728-STO-103 ~3.909 3.909 4.180 63.883 KTO-103 3.989 4.010 ( ? ) 3.969 ( ? ) 63.821 ( ? ) Not shown

PAGE 109

109 Figure 6-12. C-axis versus a-axis lattice parameters of Bi0.9Ba0.1FeO3 strained films. Figure 6-13. C/A ratio of Bi0.9Ba0.1FeO3 strained films.

PAGE 110

110 Figure 6-14. Z-contra st STEM image of Bi0.9Ba0.1FeO3 (light area) on LSAT (dark area). Figure 6-15. Z-contra st STEM image of Bi0.9Ba0.1FeO3 (light area) on LAO (dark area).

PAGE 111

111 Figure 6-16. Magnetizati on vs Field data of Bi0.9Ba0.1FeO3-x epitaxial films grown on three out of four different substrates. A) Red loop is film on LaAlO3, B) Magenta loop is film on SrTiO3, and C) Blue loop is film on LSAT. -60000-40000-200000200004000060000 -0.30 -0.20 -0.10 0.00 0.10 0.20 0.30 10K BBFO on STO M (Bper Fe)Field (Oe)-60000-40000-200000200004000060000 -0.30 -0.20 -0.10 0.00 0.10 0.20 10K BBFO on LAO M (Bper Fe)Field (Oe)-60000-40000-200000200004000060000 -0.20 -0.15 -0.10 -0.05 0.00 0.05 0.10 0.15 0.20 10K BBFO on LSAT M (Bper Fe)Field (Oe)C

PAGE 112

112 CHAPTER 7 CONCLUSION Several perovskite oxides are currently used in m any applications, su ch as transducers and multiple state memory devices due to basic scien ce studies of these materials. Realizing the importance of synthesis and i nvestigation of fundamental properties of perovskite oxide materials is needed to further pu sh advances in information stor age technology. This work is on the investigation of BaFeO3, its complex structure and magnetic properties. In addition, the synthesis and properties of a novel solid solution composed of BaFeO3 and BiFeO3 was studied. Early in this work, BaFeO3-x thin films were grown on single crystal perovskite substrates. It is found that as-deposited films were ps eudo-cubic perovskite, nonconductive and weakly ferromagnetic, which differ from that of the bulk BaFeO3. Two phases of BaFeO3 are identified, an oxygen-deficient phase and an oxygen-rich phase, where formation of either/both is dependent on growth temperature and pressure. With increase in temperature, both phases are fully visible. After annealing in oxygen, the films change from translucent to reflective and exhibit a rather large change in lattice parameter. The films also show ferromagnetism and are in some cases conductive. The resistivity is on the order of 100-200 ohm-cm. After this, further investigation of BaFeO3-x epitaxial films demonstrated complex structure, seen in reciprocal space maps. After ex situ anneal in oxygen, these complex structures clean up into a single pseudocubic st ructure that is magnetic. BaFeO3 grown under low laser fluence is not convertible via anneal to a magnetic material. Its phase composition decomposes into BaO and Fe2O3. With thickness, the crystalli ne quality of as-grown BaFeO3-x decrease, but this decrease makes it a more compa tible material to anneal and become magnetic. Artificial superlattices of BaFeO3/SrTiO3 were synthesized to attempt to decrease cracking of annealed BaFeO3, which occurs due to a large change in lattice constant when converted to a

PAGE 113

113 magnetic phase. Periodicities down to 4nm continued to result in partial cracking of the multilayers. Therefore no matter how thick or thin BaFeO3-x is, its large change in lattice constant after annealing in oxygen will present texturing (or cracking). The magnetization of ex situ annealed BaFeO3-x epitaxial films are examined as a function of direction, and was found not to have change in magnetization with direction of field with respect to crystalline plane, despite other research claims. In the end BaFeO3-x is concluded to be highly disordered, structurally and magnetically. Alloys of a solid-solution between BiFeO3 and BaFeO3-x are successfully created. The alloy films via two-target rotation method have be en carried all the way to 80% solubility. The alloy films via solid-solution targets have been successfully fabricated at near both end-memberpoints and at the half-point show ing that the solubility is possi ble over the entire range of the solid-solution. Investigation of the magnetic properties of the strained and highly crystalline Bi0.9Ba0.1FeO3 epitaxial films reveal that the sa mples are weakly ferromagnetic with a magnetization of 0.2B per Fe, more than an order of magnit ude greater than that of pure BiFeO3. Having comparable magnetic moment to BaFeO3-x after substituting roughly only 10% Ba to pure BiFeO3 is surprising and interesting. This mean s that with a small amount of Ba ions, and them being far apart from each other (and pr obably ordered structural ly), supports the claim that was made in Chapter 5 that BaFeO3-x is a highly structurally and magnetically disordered material. Theoretically, highly oxidized BaFeO3-x should have magnetic moments of more than 0.2B per Fe ion, however it does not. BaFeO3-x films are of low crystalline quality, and this supports the fact that the annealed epitaxial films also have low ma gnetic moment due to magnetic and structur al disorder

PAGE 114

114 Future challenges for the next researcher in clude investigating the ferroelectric properties of the solid solution between BaFeO3 and BiFeO3. Questions to consider include: how do the ferroelectric properties (if any) differ from that in pure BiFeO3? Also, what are the transport properties of the epitaxial films? Answers to th ese questions are keys to assessing if this solid solution has simultaneous ferroelectric and ferromagne tic behavior. Lastly, what is the effect on magnetic properties when more Ba is substituted for Bi ions? And what is the limit to this change in magnetization? The important points take away fr om this work is that BaFeO3-x, though it can be made ferromagnetic, it is a highly comp lex material. In studying BaFeO3-xs properties, conclusions can be made that its weak magnetization and unusual structure is highly disordered, magnetically and structurally. The creation of a new solid solution (Bi, Ba)FeO3 by two methods shows that a solid solution between BiFeO3 and BaFeO3-x can be synthesized. Specifi cally the creation of the alloy Bi0.9Ba0.1FeO3, shows that one can improve on BiFeO3s magnetic properties, and more importantly supports the case that BaFeO3-x exhibits magnetic and structural disorder.

PAGE 115

115 LIST OF REFERENCES 1 C. Ederer and N. A. Spaldin, Nat Mater 3, 849 (2004). 2 W. Eerenstein, N. D. Mathur, and J. F. Scott, Nature 442, 759 (2006). 3 N. A. Hill, The Journal of Physical Chemistry B 104, 6694 (2000). 4 Y. Tokura, Science 312, 1481 (2006). 5 J. F. Scott, Nat Mater 6, 256 (2007). 6 N. A. Spaldin, Magnetic materials : fundamen tals and device applications (Cambridge University Press, Cambridge, Uk; New York, 2003). 7 R. E. Hummel, Electronic properties of materials (Springer, New York, 2001). 8 S. O. Kasap, Principles of electronic materials and devices (McGraw-Hill, Boston, 2003). 9 A. Nussbaum, Electronic and magnetic behavior of materials (Prentice-Hall, Englewood Cliffs, N.J., 1967). 10 M. A. Gilleo, Physical Review 109, 777 (1958). 11 A. J. Moulson, J. M. Herbert, and I. Ne tLibrary, (Wiley, Chichester; Hoboken, NJ. 12 M. W. Barsoum and I. NetLibra ry, (Institute of Physics Pub., Bristol, UK; Philadelphia, PA. 13 K. M. Rabe, M. Dawber, C. Lichtensteiger, C. H. Ahn, and J. M. Triscone, Topics in applied physics. 105, 1 (2007). 14 M. Erchak, I. Fankuchen, and R. Ward, Jour nal of the American Chemical Society 68, 2085 (1946). 15 M. Erchak and R. Ward, Journal of the American Chemical Society 68, 2093 (1946). 16 E. Lucchini, S. Meriani, and D. Minichelli, Acta Crystallographica, Section B (Structural Crystallography and Crystal Chemistry) B29 1217 (1973). 17 H. J. V. Hook, The Journal of Physical Chemistry 68, 3786 (1964). 18 S. Mori, Journal of the Physical Society of Japan 28, 44 (1970). 19 F. Iga, Y. Nishihara, T. Katayama, K. Mura ta, and Y. Takeda, Journal of Magnetism and Magnetic Materials 104-07, 1973 (1992).

PAGE 116

116 20 E. Taketani, T. Matsui, N. Fujimura, and K. Morii, IEEE Transactions on Magnetics 40 2736 (2004). 21 T. Matsui, H. Tanaka, E. Taketani, N. Fujim ura, T. Ito, and K. Morii, Journal of the Korean Physical Society 42, 1378 (2003). 22 T. Matsui, E. Taketani, N. Fujimura, T. It o, and K. Morii, Journal of Applied Physics 93, 6993 (2003). 23 C. Michel, J.-M. Moreau, G. D. Achenbach, R. Gerson, and W. J. James, Solid State Communications 7, 701 (1969). 24 J. R. Teague, R. Gerson, and W. J. James, Solid State Communications 8, 1073 (1970). 25 V. M. Yudin, Soviet Physics-Solid State 8 (1966). 26 G. A. Smolenski, Yudin, V. M., Soviet Physics-Solid State 6, 2936 (1965). 27 J. B. Neaton, C. Ederer, U. V. Waghmare, N. A. Spaldin, and K. M. Rabe, Physical Review B (Condensed Matter and Materials Physics) 71, 14113 (2005). 28 F. Kubel and H. Schmid, Acta Crystallographica, Section B: Structural Science 46, 698 (1990). 29 D. Lebeugle, D. Colson, A. Forget, a nd M. Viret, Applie d Physics Letters 91, 22907 (2007). 30 R. Seshadri and N. A. Hill, Chemistry of Materials 13, 2892 (2001). 31 J. M. Moreau, C. Michel, R. Gerson, and W. J. James, Journal of the Physics and Chemistry of Solids 32, 1315 (1971). 32 R. T. Smith, G. D. Achenbach, R. Gerson, and W. J. James, Journal of Applied Physics 39, 70 (1968). 33 I. Sosnowska, T. P. Neumaier, and E. Steichele, Journal of Physics C: Solid State Physics 15, 4835 (1982). 34 T. Moriya, Physical Review 120, 91 (1960). 35 H. Bea, M. Bibes, S. Fusil, K. Bouzehoua ne, E. Jacquet, K. Rode, P. Bencok, and A. Barthelemy, Physical Review B (Conde nsed Matter and Materials Physics) 74, 020101 (2006). 36 D. H. Kim, H. N. Lee, M. D. Biegalski, and H. M. Christen, Applied Physics Letters 92, 012911 (2008).

PAGE 117

117 37 H. Bea, Fusil, S, Bouzehouane, K, Bibes, M, Sirena, M, Herranz, G, Jacquet, E, Contour, J.-P., Japanese Journal of Applied Physics 45, L187 (2006). 38 H. Bea, M. Gajek, M. Bibes, and A. Barthe lemy, Journal of Physics: Condensed Matter, 434221 (2008). 39 S.-W. Cheong and M. Mostovoy, Nat Mater 6, 13 (2007). 40 N. A. Spaldin and M. Fiebig, Science 309, 391 (2005). 41 D. I. Khomskii, Journal of Magnetism and Magnetic Materials 306, 1 (2006). 42 M. Fiebig, Journal of P hysics D: Applied Physics 38, R123 (2005). 43 H. Zheng, et al., Science 303, 661 (2004). 44 H. Tabata, K. Ueda, and T. Kawai, Materials Science and Engineering B 56, 140 (1998). 45 M. P. Singh, W. Prellier, L. Mechin, C. Simon, and B. Raveau, Journal of Applied Physics 99, 024105 (2006). 46 M. P. Singh, W. Prellier, L. Mechin, a nd B. Raveau, Applied Physics Letters 88, 012903 (2006). 47 P. Murugavel, P. Padhan, and W. Prellier, Applied Physics Letters 85, 4992 (2004). 48 P. Murugavel, D. Saurel, W. Prellier, C. Simon, and B. Raveau, Applied Physics Letters 85, 4424 (2004). 49 P. Murugavel, M. P. Singh, W. Prellier, B. Mercey, C. Simon, and B. Raveau, Journal of Applied Physics 97, 103914 (2005). 50 M. P. Singh, W. Prellier, C. Simon, a nd B. Raveau, Applied Physics Letters 87, 022505 (2005). 51 N. Hur, S. W. Cheong, S. N. Kale, S. B. Ogale, R. Choudhary, S. R. Shinde, and T. Venkatesan, Applie d Physics Letters 86, 112507 (2005). 52 K. Fukae, T. Takahashi, T. Yoshimura, and N. Fujimura, Ultrasonics Ferroelectrics and Frequency Control, IEEE Transactions on 55, 1056 (2008). 53 J. S. Kim, C. I. Cheon, C. H. Lee, and P. W. Jang, Journal of Applied Physics 96, 468 (2004). 54 V. A. Khomchenko, D. A. Kiselev, I. K. Bdikin, V. V. Shvartsman, P. Borisov, W. Kleemann, J. M. Vieira, and A. L. Kholkin, Applied Physics Letters 93, 262905 (2008).

PAGE 118

118 55 S.-T. Zhang, L.-H. Pang, Y. Zhang, M.-H. Lu, and Y.-F. Chen, Journal of Applied Physics 100, 114108 (2006). 56 S. R. Das, R. N. P. Choudhary, P. Bhattach arya, R. S. Katiyar, P. Dutta, A. Manivannan, and M. S. Seehra, Journal of Applied Physics 101 034104 (2007). 57 S.-T. Zhang, Y. Zhang, M.-H. Lu, C.-L. Du, Y.-F. Chen, Z.-G. Liu, Y.-Y. Zhu, N.-B. Ming, and X. Q. Pan, Applied Physics Letters 88, 162901 (2006). 58 G. L. Yuan, S. W. Or, J. M. Liu, a nd Z. G. Liu, Applied Physics Letters 89 052905 (2006). 59 C. Fanggao, S. Guilin, F. Kun, Q. Ping, and Z. Qijun, Journal of Rare Earths 24, 273 (2006). 60 Z. Quan, W. Liu, H. Hu, S. Xu, B. Sebo, G. Fang, M. Li, and X. Zhao, Journal of Applied Physics 104, 084106 (2008). 61 A. Z. Simoes, A. H. Aguiar, A. H. M. Gonzalez, J. Andres, E. Longo, and J. A. Varela, Journal of Applied Physics 104 (2008). 62 V. A. Khomchenko, et al., Journal of Physics: Condensed Matter, 155207 (2008). 63 V. R. Palkar, D. C. Kundaliya, S. K. Ma lik, and S. Bhattacharya, Physical Review B 69, 212102 (2004). 64 J. Dhahri, M. Boudard, S. Zemni, H. Rousse l, and M. Oumezzine, Journal of Solid State Chemistry 181 802 (2008). 65 V. A. Khomchenko, D. A. Kiselev, E. K. Selezneva, J. M. Vieira, A. M. L. Lopes, Y. G. Pogorelov, J. P. Araujo, and A. L. Kholkin, Materials Letters 62, 1927 (2008). 66 V. A. Khomchenko, M. Kopcewicz, A. M. L. Lo pes, Y. G. Pogorelov, J. P. Araujo, J. M. Vieira, and A. L. Kholkin, Journal of Physics D: Applied Physics, 102003 (2008). 67 B. Kundys, A. Maignan, C. Martin, N. Nguyen, and C. Simon, Applied Physics Letters 92, 112905 (2008). 68 V. A. Khomchenko, D. A. Kiselev, J. M. Vieira L. Jian, A. L. Kholkin, A. M. L. Lopes, Y. G. Pogorelov, J. P. Araujo, and M. Maglione, Journal of Applied Physics 103, 024105 (2008). 69 D. H. Wang, W. C. Goh, M. Ning, and C. K. Ong, Applied Physics Letters 88, 212907 (2006).

PAGE 119

119 70 W. Eerenstein, F. D. Morrison, F. Sher, J. L. Pr ieto, J. P. Attfield, J. F. Scott, and N. D. Mathur, Philosophical Magazine Letters 87, 249 (2007). 71 C. Wang, M. Takahashi, H. Fujino, X. Zhao, E. Kume, T. Horiuchi, and S. Sakai, Journal of Applied Physics 99, 054104 (2006). 72 J. Wang, et al., Science 299, 1719 (2003). 73 V. L. Mathe, K. K. Patankar, R. N. Patil, and C. D. Lokhande, Journal of Magnetism and Magnetic Materials 270, 380 (2004). 74 H. Bea, M. Bibes, S. Petit, J. Kreisel, and A. Barthlmy, Philosophical Magazine Letters 87, 165 (2007). 75 S. Yasui, Uchida, H, Nakaki, H, Funakubo, H, Koda, Seiichiro, Japanese Journal of Applied Physics 45, 7321 (2006). 76 X. Qi, J. Dho, R. Tomov, M. G. Blamire, and J. L. MacManus-Drisc oll, Applied Physics Letters 86, 062903 (2005). 77 D. B. Chrisey and G. K. Hubler, Pulsed laser deposition of thin films (J. Wiley, New York, 1994). 78 D. P. Norton, Materials Science and Engineering: R: Reports 43 139 (2004). 79 D. H. Lowndes, D. B. Geohegan, A. A. Puretzky, D. P. Norton, and C. M. Rouleau, Science 273, 898 (1996). 80 J. Schubert, T. Heeg, and M. Wagne r, Topics in applied physics. 106, 115 (2007). 81 J. E. Mahan, Physical vapor deposi tion of thin films (Wiley, New York, 2000). 82 S. B. Ogale and MyiLibrary, Thin films and heterostructures for oxide electronics (Springer, New York, 2005). 83 C. R. Brundle, C. A. Evans, and S. Wilson, Encyclopedia of material s characterization : surfaces, interfaces, thin films (Butterworth-Heinemann ; Manning, Boston; Greenwich, CT, 1992). 84 R. L. Fagaly, Review of Scientific Instruments 77, 101101 (2006). 85 B. D. Cullity and S. R. Stock, Elements of x-ray diffraction (Prentice Hall, Upper Saddle River, NJ, 2001). 86 J. Goldstein, Scanning electron microscopy and x-ra y microanalysis : a text for biologists, materials sc ientists, and geologists (Plenum Press, New York, 1992).

PAGE 120

120 87 D. J. Craik, Magnetic oxides (Wiley, London; New York, 1975). 88 R. Ramesh and N. A. Spaldin, Nature Materials 6, 21 (2007). 89 M. Sepliarsky, S. R. Phillpot, M. G. Stachio tti, and R. L. Migoni Journal of Applied Physics 91, 3165 (2002). 90 J. Hombo, Y. Matsumoto, and T. Kawano, Journal of Solid State Chemistry 84, 138 (1990). 91 W. W. Malinofsky and H. Kedesdy, Journa l of the American Chemical Society 76, 3090 (1954). 92 T. Matsui, H. Tanaka, N. Fujimura, T. Ito, H. Mabuchi, and K. Morii, Applied Physics Letters 81, 2764 (2002). 93 D. P. Norton, B. C. Chakoumakos, J. D. B udai, and D. H. Lowndes, Applied Physics Letters 62, 1679 (1993). 94 D. P. Norton, J. D. Budai, D. H. Lownde s, and B. C. Chakoumakos, Applied Physics Letters 65, 2869 (1994). 95 H. M. Christen, E. D. Specht, D. P. Norton, M. F. Chisholm, and L. A. Boatner, Applied Physics Letters 72, 2535 (1998). 96 H. M. Christen, L. A. Boatner, J. D. Budai, M. F. Chisholm, L. A. Gea, P. J. Marrero, and D. P. Norton, Applied Physics Letters 68, 1488 (1996). 97 K. H. Kim, D. P. Norton, J. D. Budai, M. F. Chisholm, B. C. Sales, D. K. Christen, and C. Cantoni, physica status solidi (a) 200, 346 (2003). 98 R. W. Eason, Pulsed laser deposition of thin films : applications-led growth of functional materials (Wiley, Hoboken, NJ, 2007). 99 N. Takano, T. Okita, N. Nakayama, Y. Bando, Y. Takeda, O. Yamamoto, and J. B. Goodenough, in J. Solid State Chem. ; Vol/Issue: 73:1, United States, 1988), p. Pages: 140. 100 C. Callender, D. P. Norton, R. Das, A. F. Hebard, and J. D. Budai, Applied Physics Letters 92, 012514 (2008). 101 M. Coey, Nature 430, 155 (2004). 102 M. Rajeswari, et al., Applied Physics Letters 73, 2672 (1998).

PAGE 121

121 103 N. A. Hill and K. M. Rabe, (Materials Research Society, Warrendale, PA, USA, San Francisco, CA, USA, 1999), p. 157. 104 D. Higashiyama, S. Miyasaka, N. Kida, T. Arima, and Y. Tokura, Physical Review B 70, 174405 (2004). 105 O. I. Lebedev, G. Van Tendeloo, N. Hayashi, T. Terashima, and M. Takano, Philosophical Magazine 84, 3825 (2004). 106 R. J. McQueeney, J. Ma, S. Chang, J.-Q. Yan, M. Hehlen, and F. Trouw, Physical Review Letters 98, 126402 (2007). 107 P. M. Woodward, D. E. Cox, E. Moshopoulou, A. W. Sleight, and S. Morimoto, Physical Review B 62, 844 (2000). 108 T. Takeda, Y. Yamaguchi, and H. Watanabe, Journal of the Physical Society of Japan 33, 967 (1972). 109 M. Azuma, H. Kanda, A. A. Belik, Y. Shimakawa, and M. Takano, Journal of Magnetism and Magnetic Materials 310, 1177 (2007).

PAGE 122

122 BIOGRAPHICAL SKETCH Charlee Joella Callender Bennett was born in Queens, NY, of Jam aican parents, and attended St. Gerad Majella Cat holic School from age 3 until 2nd grade. Her family moved to Kissimmee, FL where she resided until graduati on from high school. She graduated from Gateway High School with honors, and was active in marching, concert, and symphonic band for 8 years. In addition, she was involved in orga nizations such as Nati onal Honor Society where she was Secretary, National Beta Club, Nationa l Key Club, and National French Honor Society where she was Vice-President. She volunteered fo r 7 years at Give Kids the World as a gift giver, and played center-forward on the high school girls soccer team. Upon completion of high school in 1999, Charlee received many acceptances to Ivy League schools, but chose Agnes Scott College to be her college home, in Decatur/Atlanta, GA. It was there at her alma mater that she met he r husband Andrew Bennett, who was at the time a student at Devry University. Charlee was activ e in Judicial Board, Af rican and West Indian Student Association (AWISA), and was an ASC Big Si ster Mentor for six stude nts all four years. She graduated from Agnes Scott College with a B achelor of Arts degree in mathematics-physics in May 2003. Charlee enrolled with an NSF-AGEP Doctoral Fellowship to University of Florida in summer 2003 and joined the Norton Research Ox ides Group in June 2004. Charlee has completed a Master of Science in Materials Scienc e and Engineering at the University of Florida (Dec 2005). She was a Research Intern at Oa k Ridge National Laboratory (ORNL) during the last two years of her Ph.D. program. At ORNL, sh e completed thesis research in epitaxial thin film deposition via laser ablation, investigating magnetic and electric properties via XRD, SQUID, and ferroelectric technique. She also mentors under-represented youth interested in science. Charlee received her Ph.D. fr om the University of Florida on May 2, 2009.