Overview

General comment

Production covered

Blue light-emitting-diode wafer: gallium-nitride layers grown by metal-organic chemical vapor deposition on a patterned sapphire substrate, then mesa-etched and contacted. This variant represents fabrication at San'an/HC Semitek/Epistar LED epitaxy using China grid electricity.

This dataset represents operation with moderate PFC gas recovery and moderate water recycling.

Modelling choices

Support energy is allocated using 50,000 wafer starts per month. That capacity represents one fab phase or production module at a high-volume site, not the combined output of a multi-phase campus.

Gas abatement (POU + central scrubber) is allocated as a flat per-wafer facility charge, not scaled per etch/CVD pass. The epitaxy record this dataset reads states that its own water covers the cooling and the scrubbing its reactor needs, so the facility cooling tower and scrubber makeup that the support calculation charges for that tool visit is not charged again here. What remains of that makeup is a whole-wafer facility allocation, and no source divides it between the epitaxy reactor and the other steps of the flow.

Scope in detail

Operational scenario scope: the scenario changes fluorinated-gas emissions to air (SF6, CF4, C4F8, CHF3 and the CF4/COF2/SO2F2 abatement byproducts that scale on destroyed mass), the NET purchased quantity of the two gases wired to recovery on the purchase side rather than the emission side (neon, and EUV-lithography hydrogen where the process carries EUV passes - which is why those two purchase rows move between scenario cells), and fresh-water intake and wastewater. It does not change NF3, whose remote-plasma clean consumes the large majority of the charge inside the chamber, so the industry control is abatement of the small residue rather than recovery; nor the cryogenic HF/PF3 etch feeds, which are scrubbed rather than captured; nor any in-chamber reaction product. The utilisation and recovery fractions behind those statements are stated with the scenario definition that this dataset's name carries.

Manufacturing-region scope: the regional options of this family swap the electricity market this dataset draws from, and the regional supplier entry for those few upstream chemicals and gases whose supplier data is published by region. They change nothing else. Facility heating, ventilation and cooling energy stays on the region the process data was built for, so per-unit energy is the same across the regional options; water, process chemicals, gases, materials, process emissions and solid waste are the same as well. Every inventory row of these regional datasets is therefore expected to match. What the separate datasets carry is the background market each region draws on.

Technology: LED GaN 100mm, optoelectronics
Geography: Representative region: China, East China Grid (San-an Xiamen / Fujian LED epitaxy dominance).

Terms used above

PFC perfluorinated compounds; in power-supply contexts, power-factor correction
POU point of use

System boundary

The figure groups this dataset's unit processes by class. It is not a count of manufacturing steps — each process runs over as many passes as the flow requires, and those pass counts ship with the dataset.

Download this figure (SVG)

Data quality and references

Composite DQI 2.4 Good

Reliability

2.6

Completeness

2.5

Temporal

2.1

Geographic

2.0

Technological

2.8

How the score is calculated: Sources used across more manufacturing steps have more influence. Sources that define assumptions for the whole model account for 81.0% of the total weighting.

Pedigree scores follow the ecoinvent data-quality matrix: 1 is the best attainable, 5 the weakest. The composite is their aggregate.

Sampling procedure: Component vendor specifications, academic papers, industry literature
Coverage status: Partial
Pedigree-scored source files: 10 — the source records behind this dataset's manufacturing operations. Each carries the five pedigree axes above; the composite DQI aggregates them.

Technosphere inputs

14 flows. Quantities are not published; they ship with the dataset on Circa.

TMGa production Process gasg · ±30%

Source citations: DOE/PNNL PNNL-21443 Table 5-6
Upstream REEL dataset: Trimethylgallium (Ga(CH3)3)
Background dataset: Modelled by REEL; see the upstream dataset above.
Uncertainty: ±30% around the published quantity. The bounds themselves ship with the dataset on Circa.
Unit: g

TMIn production Process gasg · -29.6% / +29.1%

Source citations: DOE/PNNL PNNL-21443 Tables 5-4 and 5-6
Upstream REEL dataset: Trimethylindium (In(CH3)3)
Background dataset: Modelled by REEL; see the upstream dataset above.
Uncertainty: -29.6% / +29.1% around the published quantity. The bounds themselves ship with the dataset on Circa.
Unit: g

TMAl production Process gasg · ±30%

Source citations: DOE/PNNL PNNL-21443 Table 5-6
Upstream REEL dataset: Trimethylaluminum (Al(CH3)3)
Background dataset: Modelled by REEL; see the upstream dataset above.
Uncertainty: ±30% around the published quantity. The bounds themselves ship with the dataset on Circa.
Unit: g

Water supply (municipal) Waterm3 · ±15.1%

Derivation basis: Calculated from process-water and ultrapure-water demand across all manufacturing operations, plus facility water allocated to each finished unit. The fresh intake shown is that demand less the share reused under the water-recycling scenario this dataset assumes, and wastewater follows the same balance.

The sources below come from the manufacturing operations behind this row.
Sources (inherited): Inherited, rolled up from the contributing process steps:

DOE/PNNL PNNL-21443 Table 5-6
Upstream REEL dataset: Water Supply (Municipal)
Background dataset: Modelled by REEL; see the upstream dataset above.
Notes: Net fresh-water intake, supplied as municipal water
Uncertainty: ±15.1% around the published quantity. The bounds themselves ship with the dataset on Circa.
Unit: m3

Electricity, China (CN) ElectricitykWh · -17% / +20.7%

Derivation basis: Calculated from equipment energy across all manufacturing operations, plus facility support such as cleanroom HVAC, ultrapure water, cooling, gas abatement, and bulk gases.

The sources below come from the manufacturing operations behind this row.
Sources (inherited): Inherited, rolled up from the contributing process steps:

DOE/PNNL PNNL-21443 Table 5-4

CAE Online: Teradyne Catalyst Specifications

EE Journal: "Advantest Unveils New Ultra-High-Current Power Supply" (2024)

4Semi: TEL Precio Wafer Prober Specifications

FormFactor Summit 11000/12000 Facility Planning Guide

3D InCites: "ERS Electronic Introduces High Power Dissipation Thermal Chuck" (2023)

Abachy: "How Much Energy and Water Are Required for Wafer Fabrication" (2025)
Background data: ecoinvent 3.12
Background dataset: electricity, high voltage
Uncertainty: -17% / +20.7% around the published quantity. The bounds themselves ship with the dataset on Circa.
Unit: kWh

NH₃ Process gasg · ±30%

Source citations: DOE/PNNL PNNL-21443 Table 5-6
Background data: ecoinvent 3.12
Background dataset: ammonia, anhydrous, liquid
Uncertainty: ±30% around the published quantity. The bounds themselves ship with the dataset on Circa.
Unit: g

SiH₄ Process gasg · ±30%

Source citations: DOE/PNNL PNNL-21443 Table 5-4
Background data: ecoinvent 3.12
Background dataset: silicon tetrahydride
Uncertainty: ±30% around the published quantity. The bounds themselves ship with the dataset on Circa.
Unit: g

H₂ Process gasg · ±29.9%

Source citations: DOE/PNNL PNNL-21443 Table 5-6
Background data: ecoinvent 3.12
Background dataset: hydrogen, gaseous, low pressure
Uncertainty: ±29.9% around the published quantity. The bounds themselves ship with the dataset on Circa.
Unit: g

N₂ Process gasg · ±7.8%

Derivation basis: Most of this quantity is a facility nitrogen allocation rather than an operation-by-operation consumption: a per-wafer base rate for a reference fab capacity is rescaled to this model's assumed capacity through a sub-linear power law, and only the smaller per-lithography-pass and per-etch/CVD-pass purge terms follow the process flow. The base rate and the exponent are modelling estimates with no published derivation, so two models at the same technology node and wafer size can differ on this row by more than their process flows do. The uncertainty interval shown covers the per-pass terms; the facility base is a point estimate and carries none.

The sources for this row are listed below.
Source citations: DOE/PNNL PNNL-21443 Table 5-6
Background data: ecoinvent 3.12
Background dataset: nitrogen, liquid
Uncertainty: ±7.8% around the published quantity. The bounds themselves ship with the dataset on Circa.
Unit: g

Cl₂ Process gasg · -33.3% / +66.7%

Derivation basis: Authored GaAs/AlGaAs ICP-RIE mesa-etch recipe authored in this database for VCSEL fabrication (VCSEL fabrication literature, equipment vendor specs)

No source is attached to this row.
Source citations: Not stated
Background data: ecoinvent 3.12
Background dataset: chlorine, liquid
Uncertainty: -33.3% / +66.7% around the published quantity. The bounds themselves ship with the dataset on Circa.
Unit: g

Ar Process gasg · ±0.8%

Derivation basis: Authored GaAs/AlGaAs ICP-RIE mesa-etch recipe authored in this database for VCSEL fabrication (VCSEL fabrication literature, equipment vendor specs)

No source is attached to this row.
Source citations: Not stated
Background data: ecoinvent 3.12
Background dataset: argon, liquid
Uncertainty: ±0.8% around the published quantity. The bounds themselves ship with the dataset on Circa.
Unit: g

Compressed air Process gasg · ±33.3%

Source citations: CAE Online: Teradyne Catalyst Specifications
EE Journal: "Advantest Unveils New Ultra-High-Current Power Supply" (2024)
4Semi: TEL Precio Wafer Prober Specifications
FormFactor Summit 11000/12000 Facility Planning Guide
3D InCites: "ERS Electronic Introduces High Power Dissipation Thermal Chuck" (2023)
Abachy: "How Much Energy and Water Are Required for Wafer Fabrication" (2025)
Background data: ecoinvent 3.12
Background dataset: compressed air, 700 kPa gauge
Uncertainty: ±33.3% around the published quantity. The bounds themselves ship with the dataset on Circa.
Unit: g

O₂ Process gasg

Derivation basis: Calculated from per-operation consumption, operation counts, and manufacturing yield.

No source is attached to this row.
Source citations: Not stated
Background data: ecoinvent 3.12
Background dataset: oxygen, liquid
Uncertainty: No range defined.
Unit: g

Natural gas, burned in industrial furnace (POU abatement, low-NOx) Process chemicalMJ · -50% / +150%

Derivation basis: Calculated from per-operation consumption, operation counts, and manufacturing yield.

No source is attached to this row.
Source citations: Not stated
Background data: ecoinvent 3.12
Background dataset: heat production, natural gas, at industrial furnace low-NOx >100kW
Uncertainty: -50% / +150% around the published quantity. The bounds themselves ship with the dataset on Circa.
Unit: MJ

Outputs and waste

Wastewater Wastewaterm3 · ±15.1%

Derivation basis: Calculated from the water balance: fresh-water input minus evaporation.

The sources below come from the manufacturing operations behind this row.
Sources (inherited): Inherited, rolled up from the contributing process steps:

DOE/PNNL PNNL-21443 Table 5-6
Background data: carried, no background dataset
Background dataset: No treatment route recorded for this output.
Uncertainty: ±15.1% around the published quantity. The bounds themselves ship with the dataset on Circa.
Unit: m3

Etch byproduct scrubber sludge (III-V/nitride metal chlorides) Solid wasteg · -40% / +60%

Derivation basis: Authored GaAs/AlGaAs ICP-RIE mesa-etch byproduct row authored in this database for VCSEL fabrication

No source is attached to this row.
Source citations: Not stated
Background data: ecoinvent 3.12 treatment route
Background dataset: treatment of hazardous waste, underground deposit
Uncertainty: -40% / +60% around the published quantity. The bounds themselves ship with the dataset on Circa.
Unit: g

Chamber parts (quartz, ceramics) Solid wasteg · -60.4% / +145.3%

Source citations: patents.google.com (date not recorded). US 8,622,021 B2, "High lifetime consumable silicon nitride-silicon dioxide plasma processing components".
patents.google.com (date not recorded). US 6,838,012 B2, "Methods for etching dielectric materials".
patents.google.com (date not recorded). US 7,482,550 B2, "Quartz guard ring".
heraeus-covantics.com (date not recorded). Heraeus Covantics, "Properties of fused silica" (knowledge base).
Background data: ecoinvent 3.12 treatment route
Background dataset: treatment of inert waste, sanitary landfill
Uncertainty: -60.4% / +145.3% around the published quantity. The bounds themselves ship with the dataset on Circa.
Unit: g

Intact un-eroded sputter target body (ITO) Solid wasteg · -33.4% / +66.6%

Source citations: Methods of rejuvenating sputtering targets - H.C. Starck Inc
materion.com (date not recorded). Enhance Your Sputter Deposition Yield With The Materion Approach.
materion.com (date not recorded). Sputtering Target Recycling Solutions at Materion. Vendor web page.
Background data: ecoinvent 3.12 treatment route
Background dataset: material recovery (material dependent - typically recycled)
Uncertainty: -33.4% / +66.6% around the published quantity. The bounds themselves ship with the dataset on Circa.
Unit: g

Spent sputter targets (Al) Solid wasteg · ±44.4%

Source citations: Methods of rejuvenating sputtering targets - H.C. Starck Inc
materion.com (date not recorded). Enhance Your Sputter Deposition Yield With The Materion Approach.
materion.com (date not recorded). Sputtering Target Recycling Solutions at Materion. Vendor web page.
Background data: ecoinvent 3.12 treatment route
Background dataset: market for aluminium scrap, new
Uncertainty: ±44.4% around the published quantity. The bounds themselves ship with the dataset on Circa.
Unit: g

Scrapped material (line yield) Solid wastem2 · -72.2% / +58.8%

Source citations: iedm23.mapyourshow.com (date not recorded). Boakes et al., "Cradle-to-gate Life Cycle Assessment of CMOS Logic Technologies", IEDM 2023, 28-1. PDF document.
escholarship.org (date not recorded). Boyd, "Life-cycle Assessment of Semiconductors" (UC Berkeley dissertation, escholarship qt8bv2s63d). PDF document.
microlab.berkeley.edu (date not recorded). Leachman & Hodges, "Benchmarking Semiconductor Manufacturing" (CSM 28-fab IRW paper). PDF document.
ieor.berkeley.edu (date not recorded). Leachman (ed.), "The Competitive Semiconductor Manufacturing Survey - Third Report on Results of the Main Phase" (CSM-52), UC Berkeley, 2002. PDF document.
Background data: ecoinvent 3.12 treatment route
Background dataset: treatment of waste electric and electronic equipment, shredding
Notes: Calculated reject material after accounting for a manufacturing yield of 90%. Wafer starts that do not complete processing, at the modeled line yield of 90 percent. Line yield here counts test, monitor, and damaged wafers against wafers processed - the inclusive convention of the cited fab benchmarking and life cycle studies. The 0.85 to 0.97 uncertainty band on the yield spans measured multi-fab benchmarking: the low end reflects below-average lines and deep multi-layer flows; the high end reflects the best benchmarked lines, which reach about 97 to 98 percent.
Uncertainty: -72.2% / +58.8% around the published quantity. The bounds themselves ship with the dataset on Circa.
Unit: m2

Emissions to air

5 elementary flows released to air by this dataset's own operations. Quantities are not published; they ship with the dataset on Circa.

In(CH₃)₃ Emission to airg

Derivation basis: No corresponding ecoinvent dataset or flow was identified for this entry, so it is published with its own quantity and no background link

The sources for this row are listed below.
Source citations: DOE/PNNL PNNL-21443 Tables 5-4 and 5-6
Compartment: Air (non-urban air or from high stacks)
CAS number: 3385-78-2
ecoinvent 3.12 elementary flow: Not stated
Uncertainty: A minimum-maximum range is defined for this flow. Bounds ship with the dataset on Circa.
Unit: g

NH₃ Emission to airg

Source citations: DOE/PNNL PNNL-21443 Table 5-6
Compartment: Air (non-urban air or from high stacks)
CAS number: 7664-41-7
ecoinvent 3.12 elementary flow: Ammonia
Uncertainty: A minimum-maximum range is defined for this flow. Bounds ship with the dataset on Circa.
Unit: g

Mg(C₅H₅)₂ Emission to airg

Derivation basis: No corresponding ecoinvent dataset or flow was identified for this entry, so it is published with its own quantity and no background link

The sources for this row are listed below.
Source citations: arXiv:2203.10654
Compartment: Air (non-urban air or from high stacks)
CAS number: 1284-72-8
ecoinvent 3.12 elementary flow: Not stated
Uncertainty: A minimum-maximum range is defined for this flow. Bounds ship with the dataset on Circa.
Unit: g

Hydrogen chloride Emission to airg

Derivation basis: Authored GaAs/AlGaAs ICP-RIE mesa-etch recipe authored in this database for VCSEL fabrication (VCSEL fabrication literature, equipment vendor specs)

No source is attached to this row.
Source citations: Not stated
Compartment: Air (non-urban air or from high stacks)
Formula: HCl
CAS number: 7647-01-0
ecoinvent 3.12 elementary flow: Hydrochloric acid
Notes: BCl3 + 3H2O → B(OH)3 + 3HCl; air HCl = stoich×0.02 scrubber slip (cross-compartment split)
Uncertainty: A minimum-maximum range is defined for this flow. Bounds ship with the dataset on Circa.
Unit: g

Cl₂ Emission to airg

Derivation basis: Authored GaAs/AlGaAs ICP-RIE mesa-etch recipe authored in this database for VCSEL fabrication (VCSEL fabrication literature, equipment vendor specs)

No source is attached to this row.
Source citations: Not stated
Compartment: Air (non-urban air or from high stacks)
CAS number: 7782-50-5
ecoinvent 3.12 elementary flow: Chlorine
Uncertainty: A minimum-maximum range is defined for this flow. Bounds ship with the dataset on Circa.
Unit: g

Emissions to water

1 elementary flow released to water by this dataset's own operations. Quantities are not published; they ship with the dataset on Circa.

Chloride Emission to waterg

Derivation basis: Authored GaAs/AlGaAs ICP-RIE mesa-etch recipe authored in this database for VCSEL fabrication (VCSEL fabrication literature, equipment vendor specs)

No source is attached to this row.
Source citations: Not stated
Compartment: Water (surface water)
Formula: Cl-
CAS number: 16887-00-6
ecoinvent 3.12 elementary flow: Chloride
Notes: BCl3 scrubbed Cl⁻ to wastewater (98% POU scrubber; Cl⁻/HCl mass ratio 35.45/36.46)
Uncertainty: A minimum-maximum range is defined for this flow. Bounds ship with the dataset on Circa.
Unit: g

Flows not quantified

5 flows are recorded for this dataset but carry no quantity — cut off below the significance threshold, or with no background dataset available. They remain inside the declared system boundary; each is listed with its reason.

Cut off - below the significance threshold (4)

Sapphire (Al₂O₃) substrate Substrate materialm2 · -7.2% / +5.9%

Derivation basis: Calculated from per-operation consumption, operation counts, and manufacturing yield.

No source is attached to this row.
Source citations: Not stated
Background data: cut off
Background dataset: Not applicable: this flow is not quantified in the inventory.
Notes: Sapphire (Al2O3) substrate (100mm diameter, 0.007854 m2)
Reason: No wafer-level production dataset available in ecoinvent for Sapphire (Al2O3). Raw material datasets (e.g. gallium, silicon carbide powder) exist but miss crystal growth and wafering which dominate the environmental footprint.
Uncertainty: -7.2% / +5.9% around the published quantity. The bounds themselves ship with the dataset on Circa.
Unit: m2

ITO sputtering target Materialkg · -28.6% / +42.9%

Derivation basis: This flow is a consumable of the manufacturing operation. The model reads it from the operation's own inventory rather than from the product's bill of materials.

The sources for this row are listed below.
Source citations: freepatentsonline.com (date not recorded). US 6,852,202 B2 - Applied Materials, small planar magnetron with non-uniform erosion. Patent.
freepatentsonline.com (date not recorded). US 2007/0068804 A1 - TSMC, PVD target end-of-service-life determination. Patent.
Methods of rejuvenating sputtering targets - H.C. Starck Inc
materion.com (date not recorded). Enhance Your Sputter Deposition Yield With The Materion Approach.
Background data: cut off
Background dataset: Not applicable: this flow is not quantified in the inventory.
Reason: The sputtering target is recorded at its authored mass. No representative production dataset for the target metal is available, so its upstream burden is left unlinked rather than approximated with an unrelated metal.
Uncertainty: -28.6% / +42.9% around the published quantity. The bounds themselves ship with the dataset on Circa.
Unit: kg

Al sputtering target Materialkg · -7.1% / +28.6%

Derivation basis: This flow is a consumable of the manufacturing operation. The model reads it from the operation's own inventory rather than from the product's bill of materials.

The sources for this row are listed below.
Source citations: freepatentsonline.com (date not recorded). US 6,852,202 B2 - Applied Materials, small planar magnetron with non-uniform erosion. Patent.
freepatentsonline.com (date not recorded). US 2007/0068804 A1 - TSMC, PVD target end-of-service-life determination. Patent.
Methods of rejuvenating sputtering targets - H.C. Starck Inc
materion.com (date not recorded). Enhance Your Sputter Deposition Yield With The Materion Approach.
Background data: cut off
Background dataset: Not applicable: this flow is not quantified in the inventory.
Reason: The sputtering target is recorded at its authored mass. No representative production dataset for the target metal is available, so its upstream burden is left unlinked rather than approximated with an unrelated metal.
Uncertainty: -7.1% / +28.6% around the published quantity. The bounds themselves ship with the dataset on Circa.
Unit: kg

BCl₃ Process gasg · -40% / +60%

Derivation basis: Authored GaAs/AlGaAs ICP-RIE mesa-etch recipe authored in this database for VCSEL fabrication (VCSEL fabrication literature, equipment vendor specs)

No source is attached to this row.
Source citations: Not stated
Background data: cut off
Background dataset: Not applicable: this flow is not quantified in the inventory.
Reason: Boron trichloride is consumed at low rate in metal etch, and no production dataset for it exists in the background database used for linkage. It is scrubbed in wet abatement and its primary emission product, hydrogen chloride, is tracked separately in this inventory, so no upstream production dataset is attached here.
Uncertainty: -40% / +60% around the published quantity. The bounds themselves ship with the dataset on Circa.
Unit: g

Not quantified - no background dataset available (1)

Mg(C₅H₅)₂ Process gasg · -47.4% / +163.2%

Source citations: arXiv:2203.10654
Background data: no background dataset
Background dataset: Not applicable: this flow is not quantified in the inventory.
Reason: The magnesium dopant source for the p-type layer of a gallium nitride light-emitting diode. No production activity for it exists in the background database used for linkage, and no unreviewed precursor is put in its place. The quantity is recorded here and its upstream production is left unlinked.
Uncertainty: -47.4% / +163.2% around the published quantity. The bounds themselves ship with the dataset on Circa.
Unit: g

Limitations and unquantified flows (2)

Limits this dataset declares about itself: first any limit stated in its own description, then any limit it declares flow by flow, grouped by channel. Each entry below is the model's own disclosure.

From the dataset description

At most the whole of the cooling tower and scrubber makeup remaining in the support charge is an overlap with the epitaxy record's own water; how much of it actually is remains unresolved, and the two are published as they stand.

General

Wafer-dicing kerf and filter-cake outputs on this dataset are carried unchanged from the dicing process file's reference wafer, which is a larger-diameter silicon wafer than the one modelled here, and are not rescaled to this wafer's diameter, thickness or substrate. That file publishes no wafer thickness, no substrate density and no cut length to rescale from, and states its kerf width three different ways, so no consumer-specific value can be derived from it; the filter cake is a wastewater-treatment residue with no geometric basis at all. Treat both rows as the reference wafer's figures rather than this product's. The flow name on this dataset has been localised to this substrate, but the quantity behind it has not: it is still the reference silicon wafer's.

The flows above are this dataset's own records. What follows are the scope rules set once for the whole database in the methodology report, repeated here so every dataset page carries them.

Database-wide boundary policy — applies to every REEL dataset

These boundaries are set once for the whole database, in Chapter 2 of the methodology report, and apply to this dataset wherever they are relevant to it. The excluded flows listed above are specific to this dataset.

Use phase: Product operation is outside the cradle-to-gate scope.
End-of-life treatment: Recycling and disposal are outside the cradle-to-gate scope.
Distribution and retail: The gate is a finished component ready for integration into a higher-level assembly.
Inbound transport of raw materials: Transport of purchased raw materials to the manufacturing facility is already inside the upstream "market for" datasets that users link to a background database, so it is not modelled a second time here. This does not cover freight between REEL production stages, which is modelled where a dataset authors it.
Returnable shipping containers: Where freight between production stages is modelled, the mass moved is the product itself. The shipping container (FOSB, SEMI M31) is returnable capital equipment whose per-trip share is unsourced, so its tare is excluded from the transport effort.
Photomask fabrication: A mask set's embodied burden is amortised across a high-volume production run and is not attributed per wafer. Users assessing low-volume production should add mask fabrication separately; the methodology report gives the basis for the exclusion.
Employee transport, administration and R&D overhead: Employee transportation, facility administration and R&D/pilot-production overhead are outside scope.
Capital goods: Manufacturing equipment, cleanroom construction and facility infrastructure are excluded, on the grounds of absent public data on equipment embodied energy, uncertainty in equipment lifetime and allocation, common practice in electronics LCA, and a focus on the operational inventory. Future versions may include capital goods when sufficient public data becomes available.
Precious metal recovery credits: Scrap recovery credits for precious metals are excluded pending data availability.
Wafer reclaim: Test wafers and scrap are outside the system boundary.

Inside the boundary, linked rather than modelled

Silicon ingot growth and wafer slicing: Inside the cradle-to-gate scope, but treated as upstream material inputs linked to background databases rather than modelled as REEL processes.
Freight between production stages: Where a dataset's product moves between REEL production stages - wafer fabrication to the packaging site, for example - that leg is authored as a transport service and linked to an ecoinvent freight activity. It is measured as a transport effort in tonne-kilometres, not as a mass. Route distances are authored per route class with a stated band; a lower bound of zero is a modelling statement that the two sites can be co-located, not a missing value.

Cut-off criteria

A flow is excluded from a process inventory when it contributes less than 1 % of the total mass of inputs to that unit process, or less than 1 % of its total energy input. The denominator is total process inputs, not product mass. That distinction matters in semiconductor manufacturing, where the input mass of water, chemicals and gases greatly exceeds the product mass, so the threshold removes only genuinely minor flows.

Included regardless of the cut-off

Perfluorocarbons (CF4, C2F6, SF6, NF3) - high GWP, EPA regulated
Heavy metals (Pb, Cd, Hg, Cr(VI)) - RoHS regulated, high toxicity
Volatile organics (photoresist solvents, PGMEA) - air quality
Precious metals (Au, Ag, Pd, Pt) - high embodied impacts
Ozone-depleting substances (legacy CFCs, HCFCs) - Montreal Protocol