High-k Dielectric for Flexible Displays using Anodically Oxidized Tantalum Jovan Trujillo Flexible Display Center 3/2/07.

High-k Dielectric for Flexible Displays using Anodically Oxidized Tantalum Jovan Trujillo Flexible Display Center 3/2/07

Flexible Display Center at Arizona State University2-March-2007 -3- Copyright © 2007 Arizona State University All Rights Reserved Anatomy of a Field Effect Transistor Substrate Gate Metal Gate Dielectric a-Si:H IMD n+ a-Si contact Source metal Drain metal

Flexible Display Center at Arizona State University2-March-2007 -6- Copyright © 2007 Arizona State University All Rights Reserved Goals of Tantalum Anodization Research  Fundamental  Understand relationship between process conditions and electrical characteristics  Develop spectroscopic ellipsometry techniques to characterize Ta 2 O 5 film and interfaces.  Applied  Use Ta 2 O 5 to improve pentacene based organic transistors This work is in collaboration with Parul Dhagat  Identify problems with implementing in main 6” process line. This work is in collaboration with entire process team  Developmental  Identify factors in quality control  Supply etch engineer with material  Propose approach for increasing production

Flexible Display Center at Arizona State University2-March-2007 -7- Copyright © 2007 Arizona State University All Rights Reserved How Dielectrics Work  Applied electric field causes opposing internal electric field.  Charge builds up while under voltage.  Defects in film cause charge to leak through.  Current is released when voltage removed.

Flexible Display Center at Arizona State University2-March-2007 -8- Copyright © 2007 Arizona State University All Rights Reserved Why Tantalum Oxide? MaterialProcess Dielectric Constant Problems Silicon NitridePE-CVD~7 Step coverage, low-k, low breakdown voltage. Hafnium Silicate Reactive sputtering ~12 worse step coverage, stoichiometry problems, slow deposition rate Aluminum OxideReactive sputtering ~ 9same as hafnium silicate Tantalum OxideAnodic oxidation~ 28etch selectivity, mask changes

Flexible Display Center at Arizona State University2-March-2007 -9- Copyright © 2007 Arizona State University All Rights Reserved Anodic oxidation process ( a self limiting reaction ) Platinum CathodeTantalum Anode 60 mA ramp to 100 V 0.05% vol acetic acid 5.5 L water room temp. Hydrogen bubbles Final current < 0.40 mA giving current flux of 25 fA/  m 2 @ 100 V

Flexible Display Center at Arizona State University2-March-2007 -10- Copyright © 2007 Arizona State University All Rights Reserved Role of Acid  Purpose:  Increases conductivity of solution; create more ions for reaction.  Problems:  Negative ions from acid will contaminate the oxide. Higher leakage current. Lower breakdown voltage.  Why acetic acid?  Based on paper by Kalra, Katyal, and Singh, 1989.  Acetic acid caused highest breakdown voltage.  Effect of carbon contamination appears minimal.

Flexible Display Center at Arizona State University2-March-2007 -11- Copyright © 2007 Arizona State University All Rights Reserved Measuring Thickness with Ellipsometry (first paper) FESEM Oxide (nm) SE Oxide (nm) Diff (nm)Index @ 550 nm Wafer 1189.21845.22.2223 Wafer 2194183.410.62.219 Wafer 3192185.66.42.2102 Average:191.73184.337.4 StdDev:2.4111.137

Flexible Display Center at Arizona State University2-March-2007 -12- Copyright © 2007 Arizona State University All Rights Reserved Thickness and Index Uniformity Max-Min Thickness variation < 3 nmMax – Min Index variation < 0.02 11 wafer maps have been made with ellipsometry

Flexible Display Center at Arizona State University2-March-2007 -13- Copyright © 2007 Arizona State University All Rights Reserved Effect of Initial Current on Surface Roughness 60 mA process20 mA process Roughness (rms) = 0.696 nm Roughness (mean) = 0.516 nm Roughness (peak-to-valley) = 7.22 nm Roughness (rms) = 0.564 nm Roughness (mean) = 0.476 nm Roughness (peak-to-valley) = 2.99 nm Sputtered Ta metal Roughness (rms) = 0.463 nm Roughness (mean) = 0.334 nm Roughness (peak-to-valley) = 3.36 nm Thank you Hanna Heikkinen

Flexible Display Center at Arizona State University2-March-2007 -14- Copyright © 2007 Arizona State University All Rights Reserved Electrical Characterization  Expected Dieletric Constant of ~ 28  Paper by Kalra, Katyal, and Singh reported dielectric constant using 0.05 %v/v acetic acid.  Aluminum capacitors made by sputtering through stainless steel stencil using MRC-602 “King Kong”  Automated wafer maps of dielectric constant and leakage flux made using Electroglas 2001 “Famine”, LabView program, HP4284A LCR meter, and HP3457A multimeter.  2022 1 mm 2 and 1004 4 mm 2 capacitors per wafer  Data processing done using VBA scripts and Minitab.  Outliers > 3  removed to normalize data  ANOVA and Tukey’s test used to compare wafers

Flexible Display Center at Arizona State University2-March-2007 -15- Copyright © 2007 Arizona State University All Rights Reserved First Capacitor Batch Dielectric Constant 60 mA ramp to 100 V, 1:40 hours, final current 0.40 mA

Flexible Display Center at Arizona State University2-March-2007 -23- Copyright © 2007 Arizona State University All Rights Reserved Initial Conclusions  Dielectric constant verified to be ~28  Similar pattern in contour plots indicates systematic error caused by stencil.  Leakage flux variation caused by unknown factor.  Only one wafer showed acceptable leakage flux levels at 10 V.  Something is contaminating the films.

Flexible Display Center at Arizona State University2-March-2007 -26- Copyright © 2007 Arizona State University All Rights Reserved Final Conclusions  Doping level (10 ppb – 100 ppm?) contamination significantly affects leakage flux and dielectric constant.  Stay away from HCl vapors.  Coat pallet with tantalum before sputtering product.  Substrate quality affects dielectric constant.  Iron contamination increases leakage flux and creates uniformity issues.

Flexible Display Center at Arizona State University2-March-2007 -27- Copyright © 2007 Arizona State University All Rights Reserved First Attempt at Applying to Pentacene OTFT (transistor fabrication and characterization by Parul Dhagat) Pentacene on Ta 2 O 5 Pentacene on SiO 2

Flexible Display Center at Arizona State University2-March-2007 -28- Copyright © 2007 Arizona State University All Rights Reserved Interface characterization is important  Free valence shells and surface roughness reduce field effect mobility.  Interface treatment with OTS (octaldecytrichlorosilane) can improve pentacene transistor performance.  Ellipsometry can help. [Angst, David L.; Gary W. Simmons.Moisture Absorption Characteristics of Organosiloxane Self-Assembled Monolayers. Langmuir 1991, 7, 2236-2242.]

Flexible Display Center at Arizona State University2-March-2007 -29- Copyright © 2007 Arizona State University All Rights Reserved OTS on Ta 2 O 5 Model Based Approach  Literature reports high quality OTS monolayer to be ~25 Å thick.  Cauchy model used for monolayer.  Parameters for Ta 2 O 5 held constant.  Reported thicknesses in Å

Flexible Display Center at Arizona State University2-March-2007 -30- Copyright © 2007 Arizona State University All Rights Reserved OTS on Ta2O5 fingerprint approach  RAS – Reflection Anisotropy Spectroscopy  Requires bulk to be optically isotropic  Requires interface to be optically anisotropic  Technique rotates the sample and measures change in relectance for s-polarized light. RAS of OTS on SiO 2

Flexible Display Center at Arizona State University2-March-2007 -31- Copyright © 2007 Arizona State University All Rights Reserved Kinetics of Growth  Paper by Zhang, Macdonald, Sikora and Sikora, 1998 argues final current limited by barrier field at interface, not by thickness of film.  Used phosphoric acid, 99.95% pure Ta rods, not in clean room. High Field ModelPoint Defect Model Vs. Zhang, Lei; Digby D. Macdonald; Elzbieta Sikora; Janusz Sikora. On the Kinetics of Growth of Anodic Oxide Films. Journal of the Electrochemical Society 1998, 145, 3.

Flexible Display Center at Arizona State University2-March-2007 -32- Copyright © 2007 Arizona State University All Rights Reserved Pulsed Anodization  Assume anode acts like diode impeding current flow.  Use pulsed voltage to break interface barrier and pump more current.  Expect to increase film thickness and/or improve oxide stoichiometry.  Pulsed anodization controlled using relays and LabView program.

Flexible Display Center at Arizona State University2-March-2007 -33- Copyright © 2007 Arizona State University All Rights Reserved Research Plan  Papers in progress  Spectroscopic Ellipsometry of Anodized Tantalum  High Performance Pentacene Transistors using Optimized Anodized Tantalum Process (with Parul)  Reflectance Anisotropy Spectroscopy of OTS on Anodized Tantalum (with Parul)  Physical and Electrical Film Uniformity of Anodized Tantalum Films  Tantalum Pentoxide Capacitors using Pulsed Anodization Visit: http://www.public.asu.edu/~jtrujil1

Flexible Display Center at Arizona State University2-March-2007 -34- Copyright © 2007 Arizona State University All Rights Reserved Acknowledgements The FDC group: Dr. Gregory Raupp Shawn O’Rourke Curtis D. Moyer Dirk Bottesch Barry O’Brien Edward Bawolek Michael Marrs Scott Ageno Ke Long Consuelo Romero Diane Carrillo Virginia Woolf Susan Allen Marilyn Kyler Parul Dhagat Hanna Heikkinen Engineers at J. A. Woollam Co., Inc.: Neha Singh

High-k Dielectric for Flexible Displays using Anodically Oxidized Tantalum Jovan Trujillo Flexible Display Center 3/2/07.

Similar presentations

Presentation on theme: "High-k Dielectric for Flexible Displays using Anodically Oxidized Tantalum Jovan Trujillo Flexible Display Center 3/2/07."— Presentation transcript:

Similar presentations

About project

Feedback

Log in

Auth with social network:

High-k Dielectric for Flexible Displays using Anodically Oxidized Tantalum Jovan Trujillo Flexible Display Center 3/2/07.

Similar presentations

Presentation on theme: "High-k Dielectric for Flexible Displays using Anodically Oxidized Tantalum Jovan Trujillo Flexible Display Center 3/2/07."— Presentation transcript:

Similar presentations

About project

Feedback