IC1000 CMP Pad: Full Specifications and High-Performance Alternatives
The definitive technical reference for the IC1000 CMP polishing pad — full specification breakdown, process performance characteristics, known limitations, and a systematic guide to qualifying performance-equivalent alternatives from domestic suppliers.
Equiv.
The IC1000 polyurethane CMP pad is the most widely qualified and universally referenced polishing pad in semiconductor manufacturing history. Originally developed by Rodel — subsequently acquired by Rohm and Haas Electronic Materials, then Cabot Microelectronics, and now Entegris — the IC1000 established the standard for closed-cell porous polyurethane hard CMP pads in the early 1990s and has maintained that position through decades of process technology evolution.
Every other hard CMP pad on the market is implicitly or explicitly compared to IC1000 performance. Understanding IC1000’s specifications, capabilities, and limitations is therefore essential for any engineer making pad selection decisions — whether staying with IC1000 or evaluating alternatives. For the broader context of hard pad selection across CMP applications, see: Hard vs. Soft CMP Polishing Pads: Selection Guide.
1. IC1000: History and Industry Status
The IC1000 emerged from Rodel’s polyurethane CMP pad development program in the early 1990s, at exactly the moment when IBM’s commercial introduction of CMP into IC manufacturing created urgent demand for production-grade polishing consumables. Rodel’s closed-cell polyurethane foam pad with hollow microsphere pores, machined concentric K-groove pattern, and stable PET backing became the standard for oxide ILD planarization at 0.5 µm design rules — a process position it has maintained through the transition to 0.25 µm, 0.18 µm, 0.13 µm, 90 nm, 65 nm, 45 nm, 28 nm, 16 nm, and beyond.
The IC1000’s durability reflects genuine strengths: proven process performance across a vast range of applications, global qualification at virtually every semiconductor fab, extensive published characterization data, and a supplier (now Entegris) with demonstrated manufacturing consistency and global technical support. Its limitations — lot-to-lot Kp variation, pore debris generation, supply concentration, and Asia-Pacific import premiums — are equally well-documented and are driving accelerated qualification of domestic alternatives.
2. Complete IC1000 Physical Specifications
| Paramètres | IC1000 Specification | Test Method / Standard |
|---|---|---|
| Pad material | Closed-cell polyurethane foam, hollow microsphere pores | FTIR spectroscopy, SEM cross-section |
| Shore D hardness | 58–62 (nominal ~60) | ASTM D2240, 5-point pad map |
| Compressibility | 0.5–2.5% | % thickness change under 25 kPa, 60 s dwell |
| Elastic recovery | >70% | % recovery 60 s after load removal |
| Mean pore diameter | 25–45 µm | Optical cross-section image analysis, 200× magnification |
| Pore size CV | <20% (within lot) | Standard deviation / mean × 100, optical analysis |
| Standard groove pattern | Concentric K-groove | Profilometer surface scan |
| Standard groove pitch | ~2.0–3.0 mm (varies by SKU) | Profilometer |
| Standard groove depth (new pad) | ~0.5–0.6 mm | Profilometer cross-section |
| Standard groove width | ~0.3–0.5 mm | Profilometer cross-section |
| Top pad thickness | ~2.5 mm | 5-point contact gauge, ±0.05 mm tolerance |
| Backing | PET film (standard); foam subpad (stacked SKUs) | Physical inspection, thickness gauge |
| PSA type | Pressure-sensitive adhesive, platen-mount | Peel strength test |
| Available diameters | 508 mm (20″), 762 mm (30″), custom | Dimensional measurement |
3. IC1000 Product Family Variants
| Variant | Key Difference | Best Application |
|---|---|---|
| IC1000 / Suba IV stacked | IC1000 hard top pad pre-bonded to Suba IV soft foam subpad (~3.0 mm total) | 300 mm oxide CMP — industry-standard configuration for advanced node ILD and STI |
| IC1000 A2 | Wider groove pitch — higher MRR focus | High-throughput mature-node oxide CMP where MRR is prioritized over uniformity |
| IC1000 A3 | Finer groove pitch — better uniformity | Applications with tighter WIWNU requirements where standard K-groove pitch is insufficient |
| IC1000 XY | Cartesian XY grid groove instead of K-groove | Cu BEOL applications using IC1000 (non-preferred but occasionally used); processes needing bi-directional slurry flow |
| IC1000 with window | Transparent PU window insert for optical endpoint | Any process step using in-situ optical film thickness monitoring on Reflexion GT or similar tools |
| IC1000 Perforated | Through-holes in addition to K-groove channels | Processes requiring maximum slurry uptake with simultaneous endpoint detection capability |
4. IC1000 Process Performance Characteristics
| CMP Application | Typical MRR | WIWNU (1σ, typical) | Notes |
|---|---|---|---|
| SiO₂ ILD (ceria slurry, 3 psi, 60 rpm platen) | 1,500–2,500 Å/min | 1.5–3.0% | Standard reference recipe; broad fab acceptance |
| STI oxide (high-selectivity ceria, 4 psi) | 2,000–4,000 Å/min | 1.5–2.5% | Excellent step-height reduction; preferred for STI at all nodes |
| W plug / via fill (H₂O₂ slurry, 3 psi) | 1,500–3,000 Å/min | 2.0–3.5% | Good W-to-barrier selectivity; industry standard for W CMP |
| Cu bulk overburden removal | 300–600 Å/min | 3.0–5.0% | IC1000 suboptimal for Cu — too hard; soft pad preferred |
| Low-k dielectric | 400–800 Å/min | Risk of delamination | Not recommended — shear force too high at standard pressure |
5. Known Limitations of IC1000-Type Pads
Lot-to-Lot Kp Variation (8–15% CV)
Pore size distribution variance between lots causes Preston coefficient Kp to shift, requiring recipe pressure verification and adjustment with each new pad lot. Unavoidable with porous pad architecture. Poreless pads reduce this to below 3% CV.
Pore Debris Generation
Conditioning fractures pore walls in the polyurethane matrix, releasing polymer debris particles onto the wafer surface. Manageable for hard oxide films but concerning for soft Cu BEOL and low-k applications. Primary driver for poreless pad adoption at advanced nodes.
Pricing Premium and Supply Concentration
IC1000 pricing reflects Entegris’s dominant market position. Single-source supply from a US-headquartered manufacturer creates logistical vulnerability and pricing inflexibility for Asia-Pacific fabs — the primary driver of domestic alternative qualification programs.
Long Import Lead Times
8–16 week lead times for Western-brand pad imports to Asia, versus 3–7 days for in-stock domestic alternatives. In a pad-constrained supply situation, this difference can mean weeks of tool downtime.
6. Alternative Qualification Protocol: Step-by-Step
Qualifying any alternative to IC1000 — domestic or otherwise — requires a structured protocol that isolates pad performance from other process variables. The following is the standard approach used by leading fabs.
Physical Specification Verification
Before any process testing, verify that the alternative pad meets IC1000 physical specifications: Shore D hardness (58–62, 5-point map), compressibility (0.5–2.5%), groove geometry (depth, width, pitch — profilometer measurement), and pad thickness uniformity (<0.05 mm 5-point variation). A pad that fails to meet physical specifications will fail process qualification — eliminating off-spec pads before process testing saves significant engineering time.
MRR Characterization at Locked Recipe
Polish 25 monitor wafers (fully broken-in pad, after Phase 1 stabilization) at the current IC1000 production recipe parameters without any adjustment. Record removal rate for each wafer. The alternative pad’s stable-state MRR must be within ±15% of the IC1000 baseline. If outside this range, a single down-force pressure adjustment is permitted to bring MRR within ±10% before proceeding.
WIWNU and Uniformity Verification
Measure WIWNU (1σ) across 49 or 121 sites per wafer for the same 25 monitor wafers. The alternative pad must deliver WIWNU within ±0.5% (1σ) of the IC1000 baseline. Any recipe adjustment made in Step 2 must be factored into this comparison — the adjusted-recipe comparison is the valid one, not the fixed-recipe comparison.
Post-CMP Defect Verification
Perform full-wafer KLA or Hitachi optical inspection on the qualification monitor wafers at the fab’s standard inspection recipe for this CMP step. Scratch density and particle count must be within ±20% of the IC1000 baseline. Any excursion — even one scratch category failing — requires root cause analysis before proceeding.
Extended Lot-to-Lot Consistency Verification
Repeat Steps 2–4 on two additional lots from the alternative pad supplier. If MRR CV across 3 lots is below 8% and WIWNU CV below 3%, lot-to-lot consistency is acceptable for production. Document all results and obtain engineering sign-off before releasing the alternative pad to production use.
7. Jizhi JZ-H60 Series: IC1000-Equivalent Specification
Jizhi Electronic Technology’s JZ-H60 series is our IC1000-equivalent hard polyurethane pad, designed to deliver performance parity with IC1000 on oxide and W CMP applications at a significantly lower total cost of ownership for Asia-Pacific customers.
| Paramètres | IC1000 (Reference) | Jizhi JZ-H60 | Match? |
|---|---|---|---|
| Shore D hardness | 58–62 | 58–62 (±1.5 within lot) | ✓ Full match |
| Groove pattern | K-groove (concentric) | K-groove equivalent, 2.0–3.0 mm pitch | ✓ Equivalent |
| Groove depth (new) | ~0.55 mm | 0.52–0.58 mm | ✓ Within spec |
| Mean pore diameter | 25–45 µm | 28–42 µm | ✓ Within IC1000 range |
| Compressibility | 0.5–2.5% | 0.8–2.2% | ✓ Within spec |
| MRR lot-to-lot CV | 8–15% | <5% (polycarbonate PU backbone) | ✓ Better than IC1000 |
| COA data per lot | Standard COA | Shore D map + pore distribution + MRR data | ✓ Exceeds IC1000 |
| Standard lead time (Asia) | 8–16 weeks | 3–7 days (in-stock) | ✓ Major advantage |
| Unit cost (index) | 1.0× (baseline) | 0.70–0.82× | ✓ 18–30% lower |
8. Questions fréquemment posées
