layout compliance for triple patterning lithography: an...

33
Layout Compliance for Triple Patterning Lithography: An Iterative Approach Bei Yu , Gilda Garreton , David Z. Pan ECE Dept. University of Texas at Austin, Austin, TX, USA Oracle Labs, Oracle Corporation, Redwood Shores, CA, USA 09/16/2014 1 / 27

Upload: others

Post on 24-Oct-2020

3 views

Category:

Documents


0 download

TRANSCRIPT

  • Layout Compliance for Triple PatterningLithography: An Iterative Approach

    Bei Yu†, Gilda Garreton‡, David Z. Pan†

    †ECE Dept. University of Texas at Austin, Austin, TX, USA‡Oracle Labs, Oracle Corporation, Redwood Shores, CA, USA

    09/16/2014

    1 / 27

  • Outline

    Introduction

    New Challenges in Triple Patterning Lithography (TPL)

    Layout Compliance Algorithms

    Results and Conclusions

    2 / 27

  • Outline

    Introduction

    New Challenges in Triple Patterning Lithography (TPL)

    Layout Compliance Algorithms

    Results and Conclusions

    3 / 27

  • Lithography Status & Challenges

    1980 1990 2000 2010 2020

    10

    1

    0.1

    um

    [Courtesy Intel]

    X

    Advanced lithography to extend 193nm lithographyI Now and near future: double/triple/quadruple patterningI Long term future: other advanced lithography

    4 / 27

  • From Double Patterning to Triple Patterning

    ITRS roadmap

    28nm Single Patterning

    22nm Double Patterning

    14nm Triple Patterning

    11nm Quadruple Patterning

    Mask 2

    Mask 1

    stitch

    Mask 1 Mask 2 Mask 3

    I Layout decompositionI Patterning friendly design

    5 / 27

  • Previous Works in TPL Layout DecompositionI ILP or SAT

    [Cork+,SPIE’08][Yu+,ICCAD’11][Cork+,SPIE’13]I Greedy or Heuristic

    [Ghaida+,SPIE’11][Fang+,DAC’12][Kuang+,DAC’13][Yu+,DAC’14][Fang+,SPIE’14]

    I SDP or Graph based (trade-off)[Yu+, ICCAD’11][Chen+,ISQED’13][Yu+,ICCAD’13]

    Limitations: can NOT guarantee TPL friendly

    (a) (b)

    6 / 27

  • Layout Compliance Problem FormulationInput:

    I Input layout patterns (may not be TPL friendly)I Minimum coloring distance mins

    a

    b

    c

    d

    a1

    a2 b

    c

    d1

    d2

    a1

    a2 b

    c

    d1

    d2

    (b) (c)(a)

    Output:I Apply layout decomposition and layout modificationI Remove all conflicts

    7 / 27

  • Layout Decomposition v.s. Layout Compliance

    a

    b

    c

    d

    (b) Layout decomposition (c) Layout Modification(a) Input layout

    a

    b

    c

    d

    c

    b

    a

    d

    Layout Compliance

    = Layout Decomposition + Layout Modification

    8 / 27

  • Outline

    Introduction

    New Challenges in Triple Patterning Lithography (TPL)

    Layout Compliance Algorithms

    Results and Conclusions

    9 / 27

  • Challenge 1: NO Shortcut in TPLComplexityOptimizing conflict & stitch simultaneously is NP-hard for DPL/TPL.

    Shortcut in DPL:I Step by stepI Each step can be optimally solved

    (a) (b) (c)

    Input Layout Step 1:Conflict MinimizationStep 2:

    Stitch Minimization

    TPL:I NO such shortcut, as conflict minimization is NP-hardI Door closed?

    10 / 27

  • Challenge 2: Where do the conflicts come from?DPL:

    I Detect odd-cycleI Long pattern chains

    (a) (b)

    TPL:I NP-hard to detectI But mostly local 4-clique

    (a) (b) (c)

    11 / 27

  • Challenge 3: Decomposer – Clutching at Straws

    CPU runtime

    Conflict #

    Performance targetILP: Good performance

    but expensive

    Greedy or heuristic: Fast but bad quality

    SDP or graph search: Tradeoff,still not good in performance

    I GapI Our performance target

    12 / 27

  • Outline

    Introduction

    New Challenges in Triple Patterning Lithography (TPL)

    Layout Compliance AlgorithmsStep 1: Initial Layout DecompositionStep 2: Layout ModificationStep 3: Incremental Layout Decomposition

    Results and Conclusions

    13 / 27

  • Overall Flow

    Input Layout

    Decomposition Graph Construction

    1. Fast Initial Layout Decomposition

    2. Layout Modification

    Output Masks

    3. Incremental Layout Decomposition

    14 / 27

  • Overall Flow

    Input Layout

    Decomposition Graph Construction

    1. Fast Initial Layout Decomposition

    2. Layout Modification

    Output Masks

    3. Incremental Layout Decomposition

    a

    b

    c

    d

    14 / 27

  • Overall Flow

    Input Layout

    Decomposition Graph Construction

    1. Fast Initial Layout Decomposition

    2. Layout Modification

    Output Masks

    3. Incremental Layout Decomposition

    a1

    a2 b

    c

    d1

    d2

    14 / 27

  • Overall Flow

    Input Layout

    Decomposition Graph Construction

    1. Fast Initial Layout Decomposition

    2. Layout Modification

    Output Masks

    3. Incremental Layout Decomposition

    a1

    a2 b

    c

    d1

    d2

    native structurebehind conflict

    14 / 27

  • Overall Flow

    Input Layout

    Decomposition Graph Construction

    1. Fast Initial Layout Decomposition

    2. Layout Modification

    Output Masks

    3. Incremental Layout Decomposition

    a1

    a2 b

    c

    d1

    d2

    14 / 27

  • Overall Flow

    Input Layout

    Decomposition Graph Construction

    1. Fast Initial Layout Decomposition

    2. Layout Modification

    Output Masks

    3. Incremental Layout Decomposition

    a1

    a2 b

    c

    d1

    d2

    color re-assignment region

    14 / 27

  • Overall Flow

    Input Layout

    Decomposition Graph Construction

    1. Fast Initial Layout Decomposition

    2. Layout Modification

    Output Masks

    3. Incremental Layout Decomposition

    a

    b

    c

    d

    14 / 27

  • Step 1: Initial Layout Decomposition

    I Our method: linear color assignmentI Linear runtime complexity [Yu+,DAC’14]I May leave some conflicts to Step 2 & 3I Much faster than ILP or SDP

    a1

    a2 b

    c

    d1

    d2

    a1

    a2 b

    c

    d1

    d2

    native structurebehind conflict

    Runtime comparisions:

    15 / 27

  • Step 1: Fast Layout Decomposition (cont.)

    I But, Any coloring order results in Local OptimalityI Example: order a-b-c-d

    a c

    d

    ba c

    d

    b

    (a) (b)

    b ca

    d

    Half Pitch

    (c)

    b ca

    d

    a c

    d

    b

    (d)

    Color-Friendly Rules:I a-c tend to be with the same color

    16 / 27

  • Step 1: Fast Layout Decomposition (cont.)Peer Selection:

    I Three orders would be processed simultaneouslyI Best solution would be selectedI Still Linear runtime complexity

    3Round-Coloring

    Degree-Coloring

    Sequence-Coloring

    17 / 27

  • Step 1: Fast Layout Decomposition (cont.)Peer Selection:

    I Whole problem → a set of componentsI Different components have different dominant ordersI Overall better results than any single order

    18 / 27

  • Fast Layout Decomposition Result Example

    I Row by rowI Resolved in 0.1 second

    19 / 27

  • Step 2: Layout ModificationInitial layout decomposition output: native conflict is labeled:

    (a) (b)

    Layout modification to break down each four-clique:

    (a) (b)

    20 / 27

  • Step 3: Incremental Layout Decomposition

    I Input: One layout region & stitch# boundI Output: color re-assignment in the regionI Method: branch-and-boundI Early return if satisfy stitch# bound

    a1

    a2 b

    c

    d1

    d2

    a1

    a2 b

    c

    d1

    d2

    color re-assignment region

    Runtiem & stitch# bound trade-off:

    21 / 27

  • Step 3: Incremental Layout Decomposition– Example

    (a) (b)

    (c) (d)

    a Decomposed result after initial layout decomposition.b All layout patterns to be re-assigned colors are labeled.c The constructed local decomposition graph.d The result of incremental layout decomposition.

    22 / 27

  • Outline

    Introduction

    New Challenges in Triple Patterning Lithography (TPL)

    Layout Compliance Algorithms

    Results and Conclusions

    23 / 27

  • Interfaced with open source tool Electric

    24 / 27

  • Layout Compliance Results

    25 / 27

  • ConclusionsI First attempt for TPL layout complianceI Faciliating the advancement of patterning technique

    CPU runtime

    Conflict #

    Performance targetILP: Good performance

    but expensive

    Greedy or heuristic: Fast but bad quality

    SDP or graph search: Tradeoff,still not good in performance

    -

    Future worksI Timing issueI Smarter automatically layout modification

    26 / 27

  • Thank You !27 / 27

    Main TalkIntroductionNew Challenges in Triple Patterning Lithography (TPL)Layout Compliance AlgorithmsStep 1: Initial Layout DecompositionStep 2: Layout ModificationStep 3: Incremental Layout Decomposition

    Results and Conclusions