{
  "schemaVersion": 1,
  "id": "atlas-lesson-clean-planarize-surface",
  "title": "Cleaning and CMP as interface preparation",
  "category": "learning",
  "format": "Atlas lesson",
  "description": "Removes residues and topography while controlling loss, shape and defects.",
  "canonicalUrl": "https://semiaifoundry.com/learn/atlas/clean-planarize-surface/",
  "publishedDate": null,
  "revisionDate": "2026-10-07",
  "author": null,
  "topic": "Fabrication / FEOL and MOL",
  "series": null,
  "license": null,
  "rightsUrl": "https://semiaifoundry.com/ai-atlas/licenses.html",
  "version": "1.3.0",
  "status": "Published educational lesson",
  "evidenceScope": "Illustrative coefficients, defect maps and acceptance ranges are not production tolerances. A spatial association does not prove a CMP root cause.",
  "contentCoverage": "published-lesson",
  "artifactUrl": "https://semiaifoundry.com/ai-atlas/lesson-data/clean-planarize-surface-b4aee2df6719e290.json",
  "interactiveUrl": "https://semiaifoundry.com/ai-atlas/#clean-planarize-surface",
  "related": [],
  "supersedes": [],
  "supersededBy": [],
  "contentHash": "f55b502488eb034aabcdb5c6c450ac29792b50782cb40c16860098ea690c3223",
  "contentLength": 5056,
  "jsonUrl": "https://semiaifoundry.com/agents/publications/atlas-lesson-clean-planarize-surface.json",
  "markdownUrl": "https://semiaifoundry.com/agents/publications/atlas-lesson-clean-planarize-surface.md",
  "sections": [
    {
      "id": "section-1",
      "title": "Cleaning and CMP as interface preparation",
      "start": 43,
      "end": 398
    },
    {
      "id": "section-2",
      "title": "The mechanism",
      "start": 398,
      "end": 1577
    },
    {
      "id": "section-3",
      "title": "Science and assumptions",
      "start": 1577,
      "end": 1605
    },
    {
      "id": "section-4",
      "title": "Preston-type removal model",
      "start": 1605,
      "end": 2158
    },
    {
      "id": "section-5",
      "title": "Engineering example",
      "start": 2158,
      "end": 2380
    },
    {
      "id": "section-6",
      "title": "Distinctions and tradeoffs",
      "start": 2380,
      "end": 2738
    },
    {
      "id": "section-7",
      "title": "Specifications and interfaces",
      "start": 2738,
      "end": 3030
    },
    {
      "id": "section-8",
      "title": "Interfaces",
      "start": 3030,
      "end": 3209
    },
    {
      "id": "section-9",
      "title": "Check your understanding",
      "start": 3209,
      "end": 3238
    },
    {
      "id": "section-10",
      "title": "Average film thickness meets target, but bonding opens cluster over dense wiring. What is the next useful evidence?",
      "start": 3238,
      "end": 3692
    },
    {
      "id": "section-11",
      "title": "Evidence and scope",
      "start": 3692,
      "end": 3992
    },
    {
      "id": "section-12",
      "title": "Sources",
      "start": 3992,
      "end": 4574
    },
    {
      "id": "section-13",
      "title": "Continue exploring",
      "start": 4574,
      "end": 4597
    },
    {
      "id": "section-14",
      "title": "Foundations",
      "start": 4597,
      "end": 4818
    },
    {
      "id": "section-15",
      "title": "Related lessons",
      "start": 4818,
      "end": 5056
    }
  ],
  "citations": [
    {
      "title": "Explore in Atlas",
      "url": "https://semiaifoundry.com/ai-atlas/#clean-planarize-surface"
    },
    {
      "title": "imec — Wafer-to-wafer hybrid bonding: 400 nm interconnect pitch",
      "url": "https://www.imec-int.com/en/articles/wafer-wafer-hybrid-bonding-pushing-boundaries-400nm-interconnect-pitch"
    },
    {
      "title": "Lam Research — Our Processes",
      "url": "https://www.lamresearch.com/products/our-processes/"
    },
    {
      "title": "KLA — Chip Manufacturing 101",
      "url": "https://www.kla.com/wp-content/uploads/ChipManufacturing101.pdf"
    },
    {
      "title": "Published lesson data",
      "url": "https://semiaifoundry.com/ai-atlas/lesson-data/clean-planarize-surface-b4aee2df6719e290.json"
    },
    {
      "title": "Atlas releases and licensing",
      "url": "https://semiaifoundry.com/ai-atlas/licenses.html"
    },
    {
      "title": "Chemical mechanical planarization",
      "url": "https://semiaifoundry.com/learn/atlas/cmp/"
    },
    {
      "title": "Contamination and the state of a surface",
      "url": "https://semiaifoundry.com/learn/atlas/contamination-and-surface-state/"
    },
    {
      "title": "BEOL layers, damascene and vias",
      "url": "https://semiaifoundry.com/learn/atlas/beol-damascene-vias/"
    },
    {
      "title": "Hybrid bonding: surface state and alignment",
      "url": "https://semiaifoundry.com/learn/atlas/hybrid-bond-surface-alignment/"
    }
  ],
  "recordHash": "d404dcd2a7d5701f25a4734c32a5a08581cc98cfff5ec4f96a6ab16568a22120",
  "contentFormat": "text/markdown",
  "content": "Fabrication / FEOL and MOL · Atlas v1.3.0\n\n# Cleaning and CMP as interface preparation\n\nRemoves residues and topography while controlling loss, shape and defects.\n\n[Explore in Atlas](https://semiaifoundry.com/ai-atlas/#clean-planarize-surface)Nanometer device cross-sections; reference flow with alternatives\n\nLearning objectiveRelate CMP and cleaning measurements to the next required interface.\n\n## The mechanism\n\nFabrication repeatedly creates surfaces that the next operation cannot use directly. Cleaning removes selected residues or contaminants; chemical mechanical planarization combines surface chemistry and mechanical interaction to reduce topography or remove an overburden. These tasks meet at the final interface but are not interchangeable. CMP can leave particles, scratches, dishing or erosion, and a post-CMP clean must preserve the films it exposes. A wafer-average removal rate can hide a locally recessed conductor or remaining material. This lesson connects incoming topography, pad and slurry state, endpoint, local pattern density and post-clean inspection to the actual surface handed downstream.\n\n1.  Map incoming thickness and pattern density before evaluating the polishing result.\n2.  Combine material-specific surface reactions with pad contact to remove exposed material.\n3.  Track endpoint and local removal rather than relying on average elapsed time.\n4.  Clean residues with a method compatible with all newly exposed films.\n5.  Measure topography and defects at relevant sites and distinguish incoming variation from process-induced change.\n\n## Science and assumptions\n\n### Preston-type removal model\n\n$$\n\\mathrm{RR}=KPV\n$$\n\nThe local empirical removal model multiplies the Preston coefficient, pressure and relative speed.\n\nSymbols and units\n\nRR\n\nLocal material removal rate\n\nm/s\n\nK\n\nEmpirical Preston coefficient for these units\n\nPa⁻¹\n\nP\n\nLocal pressure\n\nPa\n\nV\n\nRelative sliding speed\n\nm/s\n\nRR is removal rate in m/s, P pressure in Pa, V relative speed in m/s; K then has units Pa⁻¹.\n\nAssumptions Empirical local approximation. Pad state, chemistry, material, pattern density and contact mechanics make K non-universal.\n\n## Engineering example\n\nSynthetic map: two wafers have the same average removal, but one has a locally recessed metal region. The average cannot decide bonding readiness; site-resolved topography distinguishes the cases.\n\n## Distinctions and tradeoffs\n\nHigh removal rate does not establish good planarization, low dishing or low defectivity.\n\nMore polishing can clear a residual film but deepen dishing, erode surrounding material or increase defects.\n\nPlanarity supports repeated interconnect layers and intimate hybrid-bond contact, linking a wafer operation to package yield.\n\n## Specifications and interfaces\n\nCMP responses\n\nRemoval, uniformity, dishing, erosion, defects\n\nSeparate response variables measured at named spatial scales.\n\nSurface readiness\n\nResidual film and local topography\n\nThe next layer or bond sets the requirement; no universal limit is implied.\n\n### Interfaces\n\nIncoming film/topography Input\n\nPattern density, overburden and contamination state\n\nPrepared interface Output\n\nMeasured topography and cleaned exposed materials\n\n## Check your understanding\n\n### Average film thickness meets target, but bonding opens cluster over dense wiring. What is the next useful evidence?\n\n1.  A larger die count\n2.  Local topography and surface-condition measurements tied to those sites\n3.  Only the global removal-rate average\n\nAnswer and reasoning\n\nLocal topography and surface-condition measurements tied to those sites\n\nLocal shape and chemistry can control contact even when a global thickness target is satisfied.\n\n## Evidence and scope\n\nIllustrative coefficients, defect maps and acceptance ranges are not production tolerances. A spatial association does not prove a CMP root cause.\n\nCompare nominal and local removal-rate extremes at a shared polish time; dishing, erosion and bond readiness are not predicted.\n\n## Sources\n\n-   [imec — Wafer-to-wafer hybrid bonding: 400 nm interconnect pitch](https://www.imec-int.com/en/articles/wafer-wafer-hybrid-bonding-pushing-boundaries-400nm-interconnect-pitch)\n-   [Lam Research — Our Processes](https://www.lamresearch.com/products/our-processes/)\n-   [KLA — Chip Manufacturing 101](https://www.kla.com/wp-content/uploads/ChipManufacturing101.pdf)\n\n[Published lesson data](https://semiaifoundry.com/ai-atlas/lesson-data/clean-planarize-surface-b4aee2df6719e290.json) · [Atlas releases and licensing](https://semiaifoundry.com/ai-atlas/licenses.html)\n\n## Continue exploring\n\n### Foundations\n\n-   [Chemical mechanical planarization](https://semiaifoundry.com/learn/atlas/cmp/)\n-   [Contamination and the state of a surface](https://semiaifoundry.com/learn/atlas/contamination-and-surface-state/)\n\n### Related lessons\n\n-   [BEOL layers, damascene and vias](https://semiaifoundry.com/learn/atlas/beol-damascene-vias/)\n-   [Hybrid bonding: surface state and alignment](https://semiaifoundry.com/learn/atlas/hybrid-bond-surface-alignment/)"
}