Multifunctional Nanostructure for Magnetoelectric and Spintronics Application R.S. Katiyar, M. Gomez, G. Morell, L. Fonseca, W. Otano ^ , O. Perales + , M.S. Tomar + , Y. Ishikawa, R.Palai, R. Thomas , A. Kumar, V. Makrov University of Puerto Rico, Rio Piedras, Mayaguez + , Cayey ^ , Puerto Rico. Abstract: CMOS compatible Multifunctional Materials to meet the near future demand of miniaturization of Si based technology and beyond Si, were the goals of this project. We designed and optimized multiferroic nanostructures for data storage and logic systems, due to high speed, low power consumption, radiation hardness, and low costs. Many of these devices need a stack of thin film nanostructures (superlattices and heterostructures) and therefore, major part of our efforts in this period focused on demonstrating the feasibility of fabricating multiferroic thin film heterostructures along with finding new multiferroic material at room temperature. Some of the materials screened so far showed multifunctional properties especially for spintronics and magnetoelectric applications. NEW MAGNETOELECTRIC MULTIFERROICS MULTIFERROICS INTEGRATION WITH SILICON SPINTRONIC MATERIALS As main memory?? Challenge!!! FeRAM advantages Lower power usage, Faster write speed Radiation resistance Realizing the high density, like DRAM and FLASH, FeRAM is an interesting option for “universal memory” candidate 1969 on-chip memory (volatile) 1972 off-chip memory (volatile) 1992 (non volatile) 1987 (nonvolatile) Secondary Density 256 MB 128 MB GB?? Toshiba and Fujitsu news 2009 PZT, SBT, BLT 2008 Fe-RAM Current Status FeRAM currently used in SONY PS2, Electronic power meters, automotive systems, smart cards, test instrumentation, factory automation, laser printers, security systems, and other systems that require reliable storage of data without an external power source Transistor FE Capacitor 1. In 1T-1C 3-D structured nanocapacitor can improve the density, - 3D deposition is very difficult with multi- component ferroelectric thin films!!. Top electrode Bottom Electrode FE 3D FE capacitor Currently used materials : PZT, SBT, and BLT ( Pr in the range 20 to 35μC/cm 2 ) 2. Introduce a better material with high Pr e.g. BFO- 60-150 μC/cm 2 - BFO will leak the stored information!! Ferroelectric 2D FE capacitor Si FE Metal Source Drain Gate Schematic of the IT-FeRAM + Buffer layer can solve this problem But what it should be? 40 60 80 100 120 C high C Accumluation - But ferroelectric directly on Si difficult- interdiffusion Insulating buffer disadvantage. Generate depolarization field in the ferroelectric film Increase the operation voltage by weakening the electric field across the ferroelectric layer. To overcome these disadvantages: Ferroelectric with low r insulating buffer layer with high r A High-k gate-oxide may be the ideal choice as a buffer layer o Large band gap o Thermal and electrical stability o Good interface between Si 1T-1C HIGH DENSITY FeRAM 1T Multiferroic BFO based MFIS Diode P –type Si (100) DyScO 3 BFO BiFeO 3 + High remnant polarization. - Large dielectric loss and high leakage current High band offset of DyScO 3 and Si will reduce the leakage current through BFO based MFIS structures and hence of great interest for the possible memory applications. 40 80 120 160 Capacitance (pF) MIM P-E Hysteresis was leaky MFIS showed ferroelectric hysteresis with reasonable memory window (1.7V) Data retention is not really good..Severely loose the charge after 100 s. - high leakage current BFO Dynamic FeRAM??.. Ferroelectric BNT based MFIS P –type Si (100) DyScO 3 BNT Aurivillius phase Bi 4 Ti 3 O 12 (BTO) Lead free ferroelectric, Low coercive field, Less fatigue, Low processing temperatures. Rare-earth substituted derivatives (Bi 3.25 Nd 0.75 Ti 3 O 12 ) have attracted much attention in recent years for non-volatile memory Large memory window of about 4.0V compared to 1.7 V of BFO MFIS structures showed excellent data retention compared to BFO Low leakage current compared to BFO based MFIS Improved interfacial quality between DSO/Si and DSO/BNT. Resulted Publications: 1. R. Thomas, D. K. Pradhan, R. E. Melgarejo, J. J. Saavedra-Arias, N. K. Karan, R. Palai, N. M. Murari, and R.S. Katiyar ECS Transactions 13,363 (2008). 2.R. Thomas, R.E. Melgarejo, N.M. Murari, S.P. Pavunny, R.S. Katiyar, Solid State Communications 149, 2013 (2009) 3. N. M. Murari, R. Thomas, R. S. Katiyar, J. Appl. Phys. 105, 084110 (2009) 4.N. M. Murari, R. Thomas, S. P. Pavunny, J. R. Calzada, and R. S. Katiyar, Appl. Phys. Lett. 94, 142907 (2009) 5.N. M. Murari, R. Thomas, R. E. Melgarejo, S. P. Pavunny, and R. S. Katiyar J. Appl. Phys. 106, 014103 (2009) Current size 45 nm <16 nm ~ 2015 Optics Magnetism Electronics Semiconductor host Magnetic impurity Integration of magnetic functionality with electronic and optical properties of semiconductor Magnetic impurity doped ZnO based diluted magnetic semiconductors (DMS) can serve as a source of spin-polarized electrons for the Spintronics applications (Co, Al) co-doped ZnO based DMS thin films Cu-doped ZnO based DMS thin films The interface of the Al 2 O 3 /Zn 0.99 Cu 0.01 O is epitaxial; (b) the film is nearly single crystalline and defects free Al 2 O 3 (a) Zn 0.99 Cu 0.01 O -6 -4 -2 0 2 4 6 -1.00 -0.75 -0.50 -0.25 0.00 0.25 0.50 0.75 1.00 M ( /Cu) Field (kOe) Zn 0.99 Cu 0.01 O Zn 0.97 Cu 0.03 O Zn 0.95 Cu 0.05 O T = 300 K All thin films shows ferromagnetism at 300K Maximum magnetization ~ 0.76 B/Cu in 3% Cu doped sample 100 200 300 400 500 600 7 S S E high 2 5%Cu 3%Cu Intensity (abr. units) Raman shift (cm -1 ) ZnO 1%Cu * A g E low 2 Raman spectra confirms the substitution of Cu2+ up to 3% Resulted Publications: 1. K. Samanta, P. Bhattacharya, and R. S. Katiyar, J. Appl. Phys. 105, 113929 (2009) 2. K. Samanta, P. Bhattacharya, J. G. S. Duque, W. Iwamoto, C. Rettori, P. G. Pagliuso, and R. S. Katiyar, Solid State Communications 147, 305 (2008) PZT PFN PbZr 0.53 Ti 0.47 O 3 /PbFe 2/3 W 1/3 O 3 (PZT/PFW) Electrical control for magnetization -10 -5 0 5 10 -16 -8 0 8 16 Magnetization (emu/cm 3 ) Magnetic Field (kOe) (c) (b) (a) -300 -150 0 150 300 -63 -31 0 31 63 -300 -150 0 150 300 -0.3 -0.1 0.0 0.1 0.3 Polarization ( C/cm 2 ) Electric field (kV/cm) Polarization (C/cm 2 ) Electric field (kV/cm) without field 1000 Oe 2000 Oe 3000 Oe 4000 Oe 5000 Oe Recovery after removal of field Strong ME coupling in multiferroic thin film at room temperature resulted in resulted in three polarization states Two with electric field and one with magnetic field. Magnetic hysteresis at room temperature in PFW/PZT samples for 0.2PFW (a), 0.3PFW (b), and 0.4PFW (c) Better hysteresis with 20/80 composition Polarization flop under the application of external magnetic field FE hysteresis studies under the application of external magnetic field from 0 to 0.5 T. The flopped “hysteresis” at 0.5 T is given in the inset; It indicates -1, 0 and 1 three logic state for memory applications 0.1 1 10 100 1000 0 300 600 900 1200 1500 0 300 600 900 0.85 T 0.70T 0.80T Imaginary permittivity (``) Real permittivity (`) Frequency (kHz) 0.85 T 0 T 100 200 300 400 500 600 720 960 1200 1440 1680 100 200 300 400 500 600 0.01 0.1 1kHz 10kHz 100kHz 500kHz 1MHz Tangent loss () Temperature (K) Dielectric constant () Temperature (K) 1kHz 10kHz 100kHz 500kHz 1MHz Magnetic field induced Debye Relaxation 0 450 900 1350 1800 0 150 300 450 600 1 MHz 0.60 T 0.85 T `` ` 0 T 100 Hz • High dielectric constant ~ 1450 and low dielectric loss < 0.03 from 100 to 500 K • Dielectric constant varied due to the magnetic field dependence of the relaxation peak • Dielectric relaxation was induced by applied external magnetic field above 0.6T (evident from the well defined Cole - Cole plots). • Relaxation peak shifted towards lower frequency at higher magnetic field. • Critical field (~0.50T) and relaxation saturation at ~0.92T matched well with the theoretical calculations and modified Vogel-Fulcher Equation Room temperature multiferroic PZT/PFW Superlattices -4000 -2000 0 2000 4000 -60 -40 -20 0 20 40 60 Magnetization (emu/cm 3 ) Applied field (Oe) 300 K 40 400 1E-7 1E-6 1E-5 1E-4 1E-3 Current density (A/cm 2 ) Electric field (kV/cm) 250 K 300 K 350 K 400 K PZT/PFW thin films of ~300nm thickness with 8:2 periodicity The remanent polarization is ~ 33 μC/cm 2 Very high breakdown field. At 20 V (for 300 nm films) ~ 60-100 MV/m, -600 -400 -200 0 200 400 600 -100 -67 -33 0 33 67 100 Polarization (C/cm ) Electric field (kV/cm) W. Eerenstein, N.D. mathur and J.F. Scott. Nature 442, 759, (2006); J J.F. Scott , Ashok Kumar, R. Palai., M K Singh ,R.S. Katiyar et al. JACeS 91(6), 1762, (2008). Spalding et. al., Science, Vol 309, 391-392 (2005) M. Bibes and A. Barthélémy. Nature, 7, 425 (2008) Computational Nanoferronics Laboratory Marjana Ležaić (Germany) Related Publications: 1. A. Kumar et al., J. Phys. Condens. Mat., August 29, 382204 (2009)] 2. A. Kumar et al., Applied Physics Letters, 94, 212903, (2009) 3. R. Pirc et al. Physical Review B 79, 214114 (2009) 4. A. Kumar et al. JMS, DOI 10.1007/s10853-009-3503-y The magnetization in 10% Co doped ZnO thin films reduces due to incorporation of additional carriers The decrease of magnetization may be due to the degeneracy of the donor level to the conduction band -15 -10 -5 0 5 10 15 -1.0 -0.5 0.0 0.5 1.0 0 50 100 150 200 250 300 0 6 12 18 24 30 36 0 30 60 90 120 0.0 0.3 0.6 0.9 1.2 1.5 Zn 0.9-x Co 0.1 O:Al x M (emu/cm 3 ) -1 p (10 -3 emu/mole-ZnO) -1 T (K) T (K) T = 300 K x = 0 x = 0.005 x = 0.01 x = 0.015 M ( B /Co) H (kOe) a) Al Co CW % % K 0 11.0(5) -3.5(2) 0.5 8.0(5) -1.3(2) 1.0 7.0(5) -1.5(2) 1.5 8.0(5) -1.9(2) b) 30 40 50 60 70 80 (0004) Intensity (a. u) 2 (degree) (0002) Al 2 O 3 0.5%Al:ZCO 1.0%Al:ZCO 1.5%Al:ZCO 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2 3.4 3.6 0 20 40 60 80 3.0 3.1 3.2 3.3 3.4 3.5 3.6 3.7 0 2 4 6 8 10 12 ( 2 ) (x10 9 cm -2 ) Al1.5% Al1.0% Transmission (%) Photon energy (eV) Co10% Al0.5% d-d Transitions 1.88, 2.03, 2.18 eV Films are highly c-axis oriented and free from impurity phase Optical band gap increases up to 54 meV due to Al doping; this is due to the Burstein-Moss (B-M) shift Characteristic d-d transitions confirms the substitution of Co 2+ in Zn 2+ lattice site Pb(Fe 0.5 Nb 0.5 )O 3 Pb(Fe 0.5 Ta 0.5 )O 3 Pb(Fe 0.66 W 0.33 )O 3 Pb(Zr 0.5 Ti 0.5 )O 3 T c ~ 380 K T N ~ 140-150 K T c ~ 180 K T N ~ 380 K T c ~ 310 K T N ~ 150 K T c ~ 620 K Can solid solution of PZT with PFW results in novel multiferroics at room temperature?? Low coercive field~ 400 Oe and high saturation magnetization ~ 60 emu/cm 2 were obtained Imprint in ferroelectric hysteresis either due to strain, difference in work function between to and bottom electrode and existence of Polar nano regions Acknowledgements: This work was supported by DoE (#DE-FG02-08ER46526) and partially by NSF-0531171 Grants. MIM MFIS