Cu Damascene 101 - University of Utah

Cu Damascene 101 - University of Utah

Copper Damascene Plating 1/5/06 Brandon Brooks Process Development Engineer Semitool Confidential 1 Outline Why Cu Interconnects? Damascene Process Flow Parameters Affecting Cu Interconnects Backside Clean and Bevel Etch Semitool Confidential 2 Damascene Plating? Semitool Confidential 3 Why Cu Interconnects?

Interconnect Metal Properties Al Cu W Melting Pt (C) 660 1,083 3,410 Oxidation in Air Rapid; SelfSealing Slow; Not SelfSealing Inert Crystalline

2.82 1.77 5.6 As Deposited 3.0-3.3* 1.8-2.0 8-11 Resistivity (m-cm)cm) Self-cm)Diffusion Coefficient (cm2s-cm)1) @ 100 C Coefficient of Thermal Expansion (Unit/C) Best! 2.110-20 2.110-30

2410-6 * Alloy (Si, Cu) 1710-6 Al Cu Resistivity Melting Point Thermal Expansion Electromigration Resistivity Melting Point Thermal Expansion Electromigration Semitool Confidential 4 4.310-6 Why Cu Interconnects?

Interconnect Metal Properties Al Cu Ag Etch Properties Cl & Br Plasmas Cl & Br Plasmas F & Cl Plasmas Etch Rate (/min) 5,000 500 5,000 Cu has a very slow etch rate Cu halides are solid at normal temperatures

Changing from Al to Cu interconnects requires new process flow Enter Damascene plating Semitool Confidential 5 Damascene Process Flow Typical Damascene Process Flow 1. Dielectric Deposition 2. Photoresist Deposition 3. UV Exposure Todays Main Topics 4. Develop Photoresist 5. Etch Dielectric 6. Remove Photoresist 7. Barrier Deposition 8. Seed Layer Deposition 9. Electrochemical Deposition (ECD) 10. Backside Clean and Bevel Etch 11. Anneal 12. Chemical Mechanical Polish (CMP) 13. Repeat Steps 1-10 for Every Metal Layer Semitool Confidential

6 Damascene Process Flow Semitool Confidential 7 Copper Interconnect Parameters Key Factors Affecting Cu Interconnect Performance 1. 2. 3. 4. Gap-cm)Fill CD Uniformity Overburden Anneal AMDs 9 Cu Levels Semitool Confidential 8

Copper Interconnect Parameters: Gap-cm)Fill Key Parameters for Gap-cm)Fill 1. Seed and Barrier Layers 1. 2. Uniformity Thickness 2. Plating Recipe 1. 2. 3. Hot Start (Initiation) Fill Current Density Waveform 3. Plating Chemistry 1. 2. Inorganic

Organic 0.12m, 8.3:1AR Trenches Semitool Confidential 9 Copper Interconnect Parameters: Gap-cm)Fill Seed and Barrier Layers Physical Vapor Deposition (PVD) Effects Semitool Confidential 10 Copper Interconnect Parameters: Gap-cm)Fill Seed and Barrier Layer Uniformity Edge Shadowing Semitool Confidential Optimized Seed Layer

11 Copper Interconnect Parameters: Gap-cm)Fill Seed and Barrier Layer Thickness 0.30micron, 4.8:1 AR Vias 1500 Total Seed Thickness Semitool Confidential 0.30micron, 4.8:1 AR Vias 2000 Total Seed Thickness 12 Copper Interconnect Parameters: Gap-cm)Fill Plating Recipe Hot Start 2X Fill Rate on the 2V Hot Start No Hot Start

2V Hot Start 0.180 m Line Width Trenches 48 Coulombs ECD Semitool Confidential 13 Copper Interconnect Parameters: Gap-cm)Fill Plating Recipe Current Density The Effect of Current Density upon Gap Fill Optimum Fill for feature D Good Current too Low 0.35m, 4.3:1 AR Vias Gap Fill 0.35m, 4.3:1 AR Vias Optimum Current

Bad Low 0.18m, 5.1:1 AR Trench High Current Density Optimum Current Semitool Confidential 0.18m, 5.1:1 AR Trench Current too High 14 Copper Interconnect Parameters: Gap-cm)Fill Plating Recipe Waveform Waveform Cu Diffusion Additive Adsorption Bottom Up Fill Direct Current

(DC) - + 0 Pulse DC + - 0 Pulse Reverse (PR) + - 0 DC plating provides better additive adsorption Pulsed plating provides better Cu diffusion

Semitool Confidential 15 Copper Interconnect Parameters: Gap-cm)Fill Plating Chemistry Inorganic Components Organic Components 1. Copper Sulfate (CuSO4) 2. Hydrochloric Acid (HCl) 3. Sulfuric Acid (H2SO4) 1. Suppressor (PEG) 2. Accelerator (SPS) 3. Leveler (Amine) Semitool Confidential 16 Copper Interconnect Parameters: Gap-cm)Fill

Inorganic Plating Chemistry Low Copper Copper Effect on Gap Fill High Copper Semitool Confidential 17 Copper Interconnect Parameters: Gap-cm)Fill Inorganic Plating Chemistry - Cl Effect on Suppressor 20 18 Chloride Effect on Gap-cm)Fill 14 12

10 Good 8 6 4 2 0 0 50 100 150 200 Cl Concentration ppm G ap Fill CVS Stripping Peak Area (mC)

16 Bad Low High Chloride (ppm ) Semitool Confidential 18 Copper Interconnect Parameters: Gap-cm)Fill Inorganic Plating Chemistry pH 3 Acid Effect on Gap Fill Good Gap Fill pH 2 pH 2

Bad Low High Acid (g/l) Semitool Confidential 19 Copper Interconnect Parameters: Gap-cm)Fill Organic Plating Chemistry Organic Effect on Gap Fill Accelerator Catalytic effect Requires very small amount of Cl Increased current for a given potential Suppressor

Suppresses deposition Requires Cl- to adsorb onto copper surface Decreases current for a given potential Leveler Suppresses deposition at high current density areas Very low concentration (diffusion limited) Semitool Confidential 20 Copper Interconnect Parameters: Gap-cm)Fill Organic Plating Chemistry Cyclic Voltammetric Stripping Analysis (CVS) Stripping Region A A = VMS I

B = VMS + Suppressor C = VMS + Sup. & Accel. Plating Region C B V Semitool Confidential 21 Copper Interconnect Parameters: Gap-cm)Fill Better Worse Organic Plating Chemistry 0.3 High Acid 150

g/l 30 Stripping Area Stripping Area 25 0.2 Low Acid (10g/l) 0.1 Worse Better High Acid 150 g/l 0 0.01 0.02 0.03 0.04 0.05

Suppressor Concentration Semitool Confidential 15 80 g/l H2SO4 10 5 80 g/l 0 20 22 0 0 1 2

3 Accelerator Concentration 4 5 Copper Interconnect Parameters: Gap-cm)Fill Organic Plating Chemistry Semitool Confidential 23 Copper Interconnect Parameters: Gap-cm)Fill Organic Plating Chemistry Semitool Confidential 24 Copper Interconnect Parameters: Gap-cm)Fill

Organic Plating Chemistry Semitool Confidential 25 Copper Interconnect Parameters: Gap-cm)Fill Organic Plating Chemistry Semitool Confidential 26 Copper Interconnect Parameters: Gap-cm)Fill Organic Plating Chemistry Semitool Confidential 27 Copper Interconnect Parameters: Gap-cm)Fill Organic Plating Chemistry

Semitool Confidential 28 Copper Interconnect Parameters: CD Uniformity Key Parameters for Current Density Uniformity 1. Chemistry 1. 2. High Acid Low Acid 2. CFD Reactor 1. Electric Field Control Intel: 8 Cu Levels Semitool Confidential 29

Copper Interconnect Parameters: CD Uniformity Generalized Electrochemical Schematic Electrolytic Copper Deposition Ammeter V0 + Current Density = Current Surf. Area Current Path e- e- ee - - Electrolyte Cu2+ Cu2+

Surface Area Cu0 Cu2++2e-cm) Anode (Oxidation) Semitool Confidential Cu2++2e-cm) Cu0 Cathode (Reduction) 30 Copper Interconnect Parameters: CD Uniformity I edge V R elec I center Rcat

V (R elec R cat ) Cathode (Thin) Rcat 1/Bath ConductivitySeed Thickness Rcat Wafer Radius = Surface = AreaArea V Relec Relec Relec 1/Bath ConductivityBath Conductivity + Electrolyte

Ranode= 0 Semitool Confidential Anode (Thick) 31 Copper Interconnect Parameters: CD Uniformity I edge I center I EC VRcat Relec ( Relec Rcat ) How To Make I EC Small? Rcat Cathode (Thin) V Current Density Throughput V

+ Edge I Loop Rcat Relec Relec Center I Loop Electrolyte Seed Layer Thickness Wafer Radius Relec Ranode= 0 Semitool Confidential Anode (Thick) 32

Bath Conductivity Copper Interconnect Parameters: CD Uniformity Conductivity at Various Bath Conditions Conductivity (mS/cm) 600 High Acid 500 400 511 300 200 100 Low Acid 247

70 0 10 g/l H2SO4 80 g/l H2SO4 175 g/l H2SO4 50 g/Bath Conductivityl Cu 50 g/Bath Conductivityl Cu 17 g/Bath Conductivityl Cu Semitool Confidential 33 Copper Interconnect Parameters: CD Uniformity Terminal Effect 0sec Current Density

Plating Time (0,0) Semitool Confidential 5sec 15sec 30sec 60sec 120sec Wafer Radius 34 Copper Interconnect Parameters: CD Uniformity The Effect of Current Density upon Gap Fill Optimum Fill for feature D Good Current too Low

0.35mm, 4.3:1 AR Vias Gap Fill 0.35mm, 4.3:1 AR Vias Optimum Current Bad Low 0.18mm, 5.1:1 AR Trench High Current Density Optimum Current Semitool Confidential 0.18mm, 5.1:1 AR Trench Current too High 35

Copper Interconnect Parameters: CD Uniformity Are the center and edge receiving the same process? Semitool Confidential 36 Copper Interconnect Parameters: CD Uniformity Advanced Reactor Design: Multiple Anodes Robust system that can handle multiple chemistries Built for the future with the ability to handle shrinking die size Cost effective ability to handle increasing wafer diameters V1 and V2 adjusted until I EC 0 Independent of Rc and Relec Cathode + V1 +

Anode 2 Semitool Confidential Anode 1 37 V2 Copper Interconnect Parameters: CD Uniformity Conventional Reactor Virtual Anodes Electrolyte Dielectric Semitool Confidential Wafer

Physical Anodes 38 CFD Reactor Electrolyte Copper Interconnect Parameters: CD Uniformity Rotating Wafer Overflow Virtual Anode Electrolyte Bubble Trap Dielectric Concentric Annular Anodes Flow Inlet Semitool Confidential

39 Copper Interconnect Parameters: CD Uniformity Superposition of Electric Field Normalized Voltage at Cathode (V) Summed Field Anode 2 Anode 3 Anode 4 -120 -100 -80 -60 -40 -20 0

20 Wafer Diameter (mm) Semitool Confidential 40 40 60 80 Anode 1 100 120 Copper Interconnect Parameters: CD Uniformity 100 nm Seed layer, 1m deposition SEMITOOL -cm) CFD

Current Density (mA/cm^2) 511mS/Bath Conductivitycm High Acid Conventional 0sec 34 133% 30 <5% 5sec 15sec 30sec 60sec 120sec 26 22 18

Current Density (mA/cm^2) 70mS/Bath Conductivitycm Low Acid 14 34 20% 30 <5% 26 0sec 22 18 120sec 14 0

25 50 75 100 125 150 0 25 Wafer Radius (mm) Semitool Confidential 41 50 75

100 125 150 Copper Interconnect Parameters: CD Uniformity Anode Current (Amps) Dynamic Compensation for Constant Current Density 2.5 Anode 1 Anode 3 2.0 Anode 4 1.5 Anode 2 1.0 0

20 40 60 80 Deposition Time (sec) Semitool Confidential 42 100 120 Copper Interconnect Parameters: Overburden Key Parameters for Overburden A. Local Overburden (Overplating) Fill Step 1. Chemistry 1. 2.

2. 3-Component 2-Component Waveform 1. Direct Current 2. Pulse Reverse B. Global Overburden Cap Step 1. Chemistry 1. 2. 2. High Acid Low Acid CFD Reactor Semitool Confidential

43 Copper Interconnect Parameters: Local Overburden Step Up No Step Up Pulse Reverse POR Direct Current POR 2-Component Organic Package High Acid Electrolyte 3-Component Organic Package Moderate Acid Electrolyte Semitool Confidential 44 Copper Interconnect Parameters: Local Overburden Insufficient Leveler

Overplating Post-cm)CMP Residual Cu Optimized Organic Conditions No Post-cm)CMP Residual Cu Planar Deposition Semitool Confidential 45 Copper Interconnect Parameters: Global Overburden Radial control of Thickness Variation () Cu Thickness () 800 600 400 200 0 -200

-400 -600 -800 -100mm 0 Wafer Diameter (mm) Semitool Confidential 46 100 Thickness (A) Thickness (A) Copper Interconnect Parameters: Global Overburden 16,000 Raider CFD Profile Before & After 30s CMP 12,000

CFD Profile before CMP 8,000 4,000 16,000 12,000 Uniform Post-cm)CMP Profile Profile after 30s CMP POR Profile Before & After 30s CMP POR Profile before CMP 8,000 4,000 Profile after 30s CMP Wafer Diameter Semitool Confidential 47 Early Clearing!

Edge Residual! Copper Interconnect Parameters: Global Overburden Normalized Thickness CMP Profile Matching 1.1 1.08 1.06 1.04 1.02 1 0.98 0.96 -150 -100 -50 0 50

Wafer Radius (mm) ECD Profile Semitool Confidential CMP Profile 48 100 150 Copper Interconnect Parameters: Anneal Key Parameters for Anneal 1. 2. 3. Temperature Feature Size Barrier Layer Semitool Confidential

49 Copper Interconnect Parameters: Anneal Effect of Temperature As Deposited Small Grains Self Annealed Large Grains Thermally Annealed Semitool Confidential 50 Copper Interconnect Parameters: Anneal Effect of Feature Size 1.0m Trenches

0.25m Trenches Furnace Anneal Self-Anneal Semitool Confidential 51 Copper Interconnect Parameters: Anneal Effect of Barrier Layer Ta Barrier Layer Strong Surface Interaction Reduced Migration TiNx Barrier Layer Weak Surface Interaction Increased Migration Large Voids Semitool Confidential

52 Copper Interconnect Parameters: Anneal Line Resistance Optimum Anneal Condition TiNx Ta TaNx Grain Growth Optimum Anneal Temp Semitool Confidential 53 Void Formation

Backside Clean and Bevel Etch Why Backside Clean and Bevel Etch? Cu is a highly mobile ion Backside contamination can have adverse effects across the fab Unstable films on the edge of the wafer can cause surface damage at CMP Objective 1. Remove bulk Cu on the edge of the wafer 1. 2. 3. 2. Delamination Flaking Yield Problems Remove atomic Cu on the back of the wafer 1. 2. 3.

Common Photolithography Common Metrology Cu ion diffusion Semitool Confidential 54 Backside Clean and Bevel Etch Capsule 1 Chamber Cut Away Capsule 1 Features 1. 2. 3. 4. 5. Hardware control of bevel etch (BE) 0-4mm BE edge exclusion (EE) range No front side protection needed BE & backside clean simultaneously Clean N2 purged microenvironment Edge Exclusion Hardware

Semitool Confidential 55 Backside Clean and Bevel Etch Capsule Dynamics Front Side Inlet: -DI H2O -N2 Chamber Rotation Wafer Device Up Seal Back Side Inlet: -Dilute Piranha Solution -DI H2O -N2 Semitool Confidential 56

Backside Clean and Bevel Etch Capsule Dynamics Front Side Inlet: -DI H2O -N2 Chamber Rotation Wafer Device Up Seal Back Side Inlet: -Dilute Piranha Solution -DI H2O -N2 Semitool Confidential 57 Backside Clean and Bevel Etch Precision Control of Chemical Wrap-cm)Around A concentric 1.5mm EE BE clears the notch

Critical Bevel Etch Parameters 1. 2. 3. Concentricity Complete Cu Clearing Clearing the Notch Semitool Confidential 58 Backside Clean and Bevel Etch Precision Control of Concentricity Concentricity Spec (a) 0.2mm Semitool Confidential 59 Backside Clean and Bevel Etch

Precision Control of Copper Removal No Copper on Edge Exclusion Zone 52 Tilt on SEM No undercut Target ECD 1.0m E Beam Spot Magn WD 10.0kV 3500x 17.1 2.0 10 m STI. Bevel Etch 1 m ECD Copper 1.5 mm Edge Exclusion

Semitool Confidential Profilometer Reading <10 m 60 Summary Why Cu Interconnects? Resistivity Reliability Damascene Process Flow Photolithography to CMP Parameters Affecting Cu Interconnects Gap-Fill Current Density Uniformity Overburden Anneal Backside Clean and Bevel Etch Bulk Cu on the Edge Atomic Cu on the Backside

Semitool Confidential 61 Acknowledgements John Klocke Cu Damascene Group Leader Kevin Witt Cu Damascene Business Development Leader Tom Ritzdorf Director of ECD Technology Jake Cook Marketing Communications All Semitool personnel that have contributed data to this presentation Semitool Confidential 62

Recently Viewed Presentations

  • Digital Systems: Hardware Organization and Design

    Digital Systems: Hardware Organization and Design

    Computer Architecture Chapter 3: RISC vs CISC Chapter Outline Machine Characteristics and Performance RISC versus CISC Practical Aspects of Machine Cost-Effectiveness Cost for useful work is fundamental issue Mounting, case, keyboard, etc. are dominating the cost of integrated circuits Upward...
  • Systems Biology for Clinical Decision Support Systems  Outlines:

    Systems Biology for Clinical Decision Support Systems Outlines:

    Dept of Comp Med & Bioinf, Kayvan Najarian. Supervised vs. Unsupervised Learning * Supervised Learning / Classification . Training data are labeled, i.e. the output class of all training data are given. Example: predicting if a sequence is an ....
  • Ethics activity project

    Ethics activity project

    assignment- in class. 1. You will have 15 min to interview at least three people around the campus (you will be asking them the questions from the previous slide)
  • Writing Formulas Naming Compounds - BPS District

    Writing Formulas Naming Compounds - BPS District

    Naming Compounds and Writing Formulas Chapter 2 (2.6 and 2.7) Writing Names for Ionic Compounds Write the name of the cation followed by the name of the anion Ending of monatomic anions must be changed to -ide Include Roman numerals...
  • Quantum Computing A Tutorial at the 2003 Genetic

    Quantum Computing A Tutorial at the 2003 Genetic

    A Tutorial at the 2003 Genetic and Evolutionary Computation Conference (GECCO-2003) Lee Spector School of Cognitive Science Hampshire College Amherst, MA 01002, USA [email protected] Includes results of collaborations with Herbert J. Bernstein, Howard Barnum, and Nikhil Swamy. Overview What is...
  • CKDu REVISITED R.M.G. Rajapakse (Ph.D., D.I.C., C. Chem.,

    CKDu REVISITED R.M.G. Rajapakse (Ph.D., D.I.C., C. Chem.,

    Definition: Any condition that causes reduced kidney function over a period of time is a chronic kidney disease. CKD is present when a patient's glomerular filtration rate remains below 60 mL min-1 for more than 3 months, or when a...
  • Foam Application Rates for E85 and E-95 - IAHMR

    Foam Application Rates for E85 and E-95 - IAHMR

    Foam Application Rates for E85 and E-95 ... NFPA 11 states that a minimum15-minute flow duration is required. If you calculate this minimum requirement in all of your foam concentrate requirements this should get you off to a good start....
  • Sociology - Paulding County School District

    Sociology - Paulding County School District

    Sociology defined… As the study of humans in their collective aspect, sociology is concerned with all group activities—economic, social, political, and religious. Sociologists. study such areas as bureaucracy, community, deviant behavior, family, public opinion, social change, social mobility, social stratification,...