Glass & Ceramics Process Emissions
A glass furnace and a brick kiln both burn fuel and both release CO₂ from the raw materials themselves. Only the second of those is a process emission — and the most common inventory error in this sub-sector is failing to separate the two.
Get the carbonate chemistry and the combustion boundary right, and the rest of glass and ceramics accounting is arithmetic.
Glass and ceramics process CO₂ comes from carbonates (soda ash, limestone, dolomite) decomposing during melting and firing. Calculate it from carbonate input × a stoichiometric factor, or from product output × an IPCC default — never from fuel use, which is a separate combustion line.
This methodology covers the calculation of CO₂ process emissions from carbonate decomposition in glass manufacturing and ceramics firing, under IPCC 2006 Guidelines Volume 3, Chapter 2 (Mineral Industry Emissions). It sets out the two emission mechanisms, the Tier 1/2/3 calculation ladder, the cullet recycling credit, the carbonate-fraction approach for ceramics, the full stoichiometric factor set, two complete worked examples, and the edge cases and verification rules that determine whether an inventory survives audit and CBAM scrutiny.
A methodology page is the execution layer. It takes the scope concept from the glossary, the calculation rules from the IPCC mineral-industry chapter, and the stoichiometric factors from the data layer, and tells you exactly how to turn batch composition and product tonnage into a defensible process-emission inventory line. If you already know your batch recipe, jump to the glass method or the worked examples.
What Counts as a Glass/Ceramics Process Emission
Glass melting and ceramics firing each release CO₂ from two physically distinct sources, and the GHG Protocol assigns them to two different inventory categories even though both are Scope 1:
- Process emissions — CO₂ chemically liberated when carbonate raw materials decompose under heat. Soda ash, limestone and dolomite in a glass batch, or carbonate impurities in clay, break down into oxides plus CO₂. This CO₂ would be released regardless of how the heat was supplied — it is a property of the raw materials, not the fuel. This is the subject of this methodology.
- Combustion emissions — CO₂ (plus CH₄ and N₂O) from burning natural gas, fuel oil, petroleum coke or other fuel to reach melting and firing temperatures. These belong to the Scope 1 stationary combustion inventory and are calculated from fuel quantity × fuel emission factor, not from product output.
Both lines are mandatory and both are Scope 1, but they are computed from entirely different activity data and must not be conflated. Reporting a single merged “glass plant CO₂” figure derived from one method is the most frequent rejection trigger at verification, because it makes the carbonate and combustion components impossible to disaggregate — and EU ETS, CBAM and CSRD all require that disaggregation.
On a modern float-glass line, combustion CO₂ is typically larger than process CO₂ — roughly 0.28 t CO₂ per tonne of glass from fuel versus around 0.21 t CO₂ per tonne from carbonates. Misclassifying combustion fuel as a process emission (or vice versa) does not just move a number between columns; it corrupts both the process intensity benchmark and the fuel-efficiency benchmark a verifier checks against. Calculate the two lines from separate activity data — carbonate mass for process, fuel quantity for combustion — and never derive one from the other.
Provider directory
Need help keeping this up all year?
See who does this work. Every listing names the standards it works to, and paid placements are labelled. Including Pangolin Associates and Attain Zero.
Browse 6 carbon accounting & inventory providers →Do this work? A listing is US$390 a year. Get listed →
The Two Emission Mechanisms
Glass and ceramics share the same underlying chemistry — carbonate decomposition (calcination) — but the carbonates enter the process differently, which changes how the emission is calculated.
Glass — deliberate carbonate fluxes
Glass batch is engineered to contain carbonates. Soda ash (Na₂CO₃) supplies sodium oxide to lower the melting point; limestone (CaCO₃) and dolomite (CaMg(CO₃)₂) supply calcium and magnesium oxides for durability. When the batch melts at around 1,500 °C, each carbonate decomposes and releases its CO₂.
CaCO₃ → CaO + CO₂↑
CaMg(CO₃)₂ → CaO + MgO + 2 CO₂↑
Because the carbonates are deliberately added in known proportions, glass process CO₂ is highly predictable from the batch recipe — the basis for the precise Tier 3 method.
Ceramics — incidental carbonate in clay
Ceramics process CO₂ comes mostly from carbonate impurities naturally present in the raw clay — calcite and dolomite that decompose during firing — rather than from deliberately added flux. A small additional contribution can come from oxidation of organic carbon in the clay.
(plus minor organic-C oxidation)
Because the carbonate content is a property of the clay deposit rather than a recipe, ceramics CO₂ is calculated from the clay’s measured carbonate fraction, not from a product-output default — there is no reliable Tier 1 output factor for fired ceramics.
This asymmetry is why the two materials get separate calculation sections below. Glass has a usable output-based Tier 1 default and a precise input-based Tier 3 method; ceramics is computed from carbonate content with stored composition defaults, because the IPCC guidelines publish no single emission factor per tonne of fired ceramic product.
Governing Standards
Four reference layers govern glass and ceramics process accounting. They nest rather than compete: the science sets the factors, the protocol sets the boundary, and the regulations set the reporting obligation.
| Layer | Standard | What it supplies |
|---|---|---|
| Calculation method | IPCC 2006 Guidelines, Vol 3, Ch 2 | The Tier 1/2/3 hierarchy, stoichiometric carbonate emission factors, glass output defaults, and the ceramics carbonate-fraction approach. |
| Methodology refinement | IPCC 2019 Refinement | Updated default values and clarifications layered over the 2006 base method; used where a refined default supersedes the 2006 figure. |
| Inventory boundary | GHG Protocol Corporate Standard | Organisational and operational boundaries; the Scope 1 classification of both process and combustion CO₂; the requirement to disaggregate them. |
| Regulatory reporting | EU ETS · EU CBAM | Monitoring, reporting and verification obligations for installations; CBAM default factors for imported glass and ceramic goods where producer-specific data is unavailable. |
The global warming potential question is simple here: carbonate process emissions are pure CO₂, whose GWP is 1 by definition under every IPCC assessment report. Unlike fuel combustion — which carries small CH₄ and N₂O components requiring AR5 or AR6 GWP conversion — the process line needs no GWP multiplication. A tonne of process CO₂ is a tonne of CO₂e.
Choosing a Tier
The IPCC method offers three tiers of increasing precision and data demand. The choice is driven by what activity data the installation can produce and how material the source is to the inventory.
minerals.glass.tier1_default or a product-specific glass factor, then apply the cullet correction. Ceramics has no Tier 1 output factor — use the carbonate-fraction method with stored composition defaults.
Screening grade. Acceptable for first-year or immaterial sources; carries the widest uncertainty.
carbonate.* factor and sum. Tier 3 additionally applies a calcination fraction where decomposition is incomplete.
Highest accuracy for glass, since batch composition is precisely known. Preferred for material sources at audit.
carbonate.* factor. Falls back to minerals.ceramics.clay_carbonate_content_default where no site measurement exists.
Site-measured carbonate content moves ceramics from screening to audit grade.
An installation can legitimately run Tier 3 on its glass batch (where it weighs every carbonate input) and Tier 1 on a minor secondary furnace where only output is metered. The tier is chosen for each emission source on materiality and data availability, and the chosen tier is documented per source in the monitoring plan. Mixing tiers across sources is normal and accepted; what is not accepted is silently switching tiers between reporting years without restating the trend.
Glass: Calculation Method
Glass offers both an output-based Tier 1 method and a precise input-based Tier 3 method. Most installations report Tier 3 because batch composition is metered, but the Tier 1 default remains useful for screening and for cross-checking.
5.1 Tier 1 — output-based with cullet correction
The Tier 1 method multiplies the mass of glass produced by an emission factor, then reduces the result to reflect cullet (recycled glass), which has already been through calcination and releases no further process CO₂ when remelted.
minerals.glass.tier1_default = 0.2 t CO₂/t [GreenCalculus minerals.glass.tier1_default · IPCC VOL3 2006 Vol 3 Ch 2 Eq 2.13 · v2026.284], or a product-specific factor below.
The IPCC generic default for glass is minerals.glass.tier1_default = 0.2 t CO₂ per tonne [GreenCalculus minerals.glass.tier1_default] of glass. Where the product type is known, the product-specific factors below are tighter — soda-lime float and container glass run higher than the generic default, while specialty borosilicate and tableware run substantially lower because they use less carbonate flux.
| Glass type | MasterBrain key | EF (t CO₂/t glass) |
|---|---|---|
| Generic default | minerals.glass.tier1_default | 0.2 |
| Float (flat) glass | minerals.glass.float | 0.21 |
| Container glass (flint) | minerals.glass.container_flint | 0.21 |
| Container glass (amber/green) | minerals.glass.container_amber_green | 0.21 |
| E-glass fibre (reinforcement) | minerals.glass.fiberglass_e | 0.19 |
| Glass wool (insulation) | minerals.glass.fiberglass_insulation | 0.25 |
| Specialty — tableware | minerals.glass.specialty_tableware | 0.1 |
| Specialty — laboratory / pharma (borosilicate) | minerals.glass.specialty_lab_pharma | 0.03 |
Factors: IPCC 2006 Guidelines, Vol 3, Ch 2, glass sub-sector defaults, surfaced live from the MasterBrain minerals.glass.* family. Additional specialty rows (lighting, TV panel/funnel) exist in the same family.
5.2 The cullet credit
Cullet is broken or recycled glass added to the batch. Because it has already been melted once, its carbonates have already released their CO₂ — remelting cullet produces no further process emissions. The cullet credit is therefore a direct linear reduction: a batch that is 30% cullet produces 30% less process CO₂ than an all-virgin batch of the same output.
Cullet reduces process emissions in exact proportion to its batch fraction, because it removes that fraction of fresh carbonate. It also reduces combustion emissions, since remelting cullet needs less energy than calcining virgin batch — but that energy saving is a separate, smaller, non-linear effect captured in the combustion inventory, not in the (1 − fcullet) term. Do not apply a single cullet factor to a merged process-plus-combustion total; the two respond to cullet differently.
5.3 Tier 3 — carbonate inventory (the precise method)
The Tier 3 method ignores product output entirely and works from the carbonate inputs. Each carbonate consumed is multiplied by its stoichiometric factor — the exact mass of CO₂ released per tonne of that mineral — and the results are summed. Because cullet contributes no fresh carbonate, it never enters this calculation; the cullet credit is implicit.
carbonate.* family, §7 below
For a soda-lime glass, the three carbonate inputs are soda ash (carbonate.sodium_carbonate = 0.41492 t CO₂/t) [GreenCalculus carbonate.sodium_carbonate · IPCC VOL3 2006 Vol 3 Ch 2 Table 2.1], limestone (carbonate.calcite = 0.43971 t CO₂/t [GreenCalculus carbonate.calcite · IPCC VOL3 2006 Vol 3 Ch 2 Table 2.1]) and dolomite (carbonate.dolomite = 0.47732 t CO₂/t) [GreenCalculus carbonate.dolomite · IPCC VOL3 2006 Vol 3 Ch 2 Table 2.1]. Glass melting is treated as complete calcination, so F = 1.0 unless a site has documented evidence of carry-over.
A complete float-glass plant is worked end-to-end in the worked examples — both the Tier 1 output method and the Tier 3 carbonate inventory, reconciled against each other.
Ceramics: Calculation Method
Ceramics has no usable Tier 1 output factor — the IPCC guidelines publish no single emission factor per tonne of fired brick, tile or sanitaryware, because carbonate content varies far too much between clay deposits. Instead, ceramics process CO₂ is built up from the clay’s carbonate fraction.
minerals.ceramics.clay_carbonate_content_default = 0.1 [GreenCalculus minerals.ceramics.clay_carbonate_content_default · IPCC VOL3 2006 Vol 3 Ch 2 Section 2.5.1.3].
minerals.ceramics.clay_loss_factor_default = 1.1 [GreenCalculus minerals.ceramics.clay_loss_factor_default · IPCC VOL3 2006 Vol 3 Ch 2 Section 2.5.1.3].
carbonate.calcite = 0.43971 t CO₂/t) [GreenCalculus carbonate.calcite].
The two stored composition defaults — a carbonate content of 0.1 tonne [GreenCalculus minerals.ceramics.clay_carbonate_content_default] of carbonate per tonne of clay and a loss-factor multiplier of 1.1 [GreenCalculus minerals.ceramics.clay_loss_factor_default] — are screening values. They are deliberately conservative; a site that measures the actual carbonate content of its clay (by loss-on-ignition or carbon analysis) will almost always replace both with tighter, lower-uncertainty figures, which is what a reasonable-assurance audit expects for a material source.
minerals.ceramics.tier1_default
Ceramics is computed, never looked up. The MasterBrain stores the two composition inputs above but deliberately publishes no per-tonne-of-product emission factor for ceramics, because none exists in the IPCC guidance. Writing a calculation against an invented minerals.ceramics.tier1_default or minerals.bricks key would render empty and silently zero the source. Always build the ceramics figure from clay mass × carbonate fraction × loss factor × a real carbonate.* factor.
Carbonate Stoichiometry Reference
The stoichiometric emission factor for a carbonate is fixed by chemistry: it is the mass of CO₂ released per unit mass of pure carbonate when it fully decomposes, derived from molecular weights. These factors are the backbone of every input-based calculation in this methodology and are identical to IPCC 2006 Vol 3 Ch 2 Table 2.1.
| Carbonate | Formula | MasterBrain key | EF (t CO₂/t carbonate) | Where it appears |
|---|---|---|---|---|
| Calcite (limestone) | CaCO₃ | carbonate.calcite | 0.43971 | Glass flux; clay impurity; the limestone factor |
| Dolomite | CaMg(CO₃)₂ | carbonate.dolomite | 0.47732 | Glass flux supplying MgO; clay impurity |
| Sodium carbonate (soda ash) | Na₂CO₃ | carbonate.sodium_carbonate | 0.41492 | Primary glass flux (network modifier) |
| Magnesite | MgCO₃ | carbonate.magnesite | 0.52197 | Refractory ceramics; some specialty glass |
| Rhodochrosite | MnCO₃ | carbonate.rhodochrosite | 0.38286 | Colourant / decolourant carbonate |
| Siderite | FeCO₃ | carbonate.siderite | 0.37987 | Iron-bearing clay impurity |
IPCC 2006 Guidelines, Vol 3, Ch 2, Table 2.1 (stoichiometric ratios), surfaced live from the MasterBrain carbonate.* family. All values are pure CO₂; GWP = 1.
carbonate.sodium_carbonate, not a “soda_ash” alias
Two key-slug traps recur in this family. First, soda ash resolves under carbonate.sodium_carbonate — there is no carbonate.soda_ash alias, and guessing it renders an em-dash. Second, there is no carbonate.limestone or carbonate.caco3 row: limestone is calcite (CaCO₃), so use carbonate.calcite for any limestone input. Using calcite’s factor where soda ash was meant overstates the soda portion, because calcite (0.43971 [GreenCalculus carbonate.calcite]) sits above sodium carbonate (0.41492 [GreenCalculus carbonate.sodium_carbonate]) per tonne.
Worked Examples
Two complete examples computed end-to-end, with all arithmetic shown. Values are fixed — these are audit records at the snapshot date of this methodology revision, reconciling stated inputs to stated results.
8.1 Float-glass plant — Tier 1 and Tier 3 reconciled
Annual glass output:
200,000 t · Cullet fraction: 30% (0.30)Tier 3 batch (per tonne of virgin glass): soda ash
0.20 t · limestone 0.13 t · dolomite 0.13 tFurnace fuel: natural gas, ~5.5 GJ/t glass (combustion line — calculated separately, shown for boundary context only)
The two methods land within 3.7% of each other. That spread is the value of Tier 3: the output default of 0.21 t CO₂/t is a generic figure, while the carbonate inventory reflects FloatCo’s actual batch recipe and so is the defensible figure for reporting. The Tier 1 result is a legitimate cross-check that confirms no order-of-magnitude error, but the 28,307.71 t CO₂ from the carbonate inventory is what enters the inventory.
At ~5.5 GJ/t natural gas, FloatCo’s combustion CO₂ is roughly 0.28 t/t × 200,000 t ≈ 55,900 t CO₂ — larger than the 28,308 t process line. This combustion figure is calculated from gas consumption × the natural-gas fuel factor and reported on a separate Scope 1 stationary combustion line. It is shown here only to make the boundary concrete: the process methodology produces the 28,308 t figure, never the 55,900 t one.
8.2 Brick plant — ceramics carbonate-fraction method
Annual raw clay input:
150,000 tCarbonate content: screening default
0.10 t carbonate/t clay (no site measurement in year 1)Loss factor: default
1.1 · Dominant carbonate: calcite
BrickCo’s 7,255.22 t CO₂ is a screening-grade figure built entirely on the two stored composition defaults. The single largest lever to tighten it is a laboratory carbonate analysis of the clay: most structural clays sit below the 0.10 default, so a measured fraction typically reduces the reported figure and the uncertainty band together. The loss factor of 1.1 should also be revisited against measured organic-carbon content, since clays low in organics justify a lower multiplier.
Edge Cases
Five situations recur in glass and ceramics inventories and each has a defined treatment.
- Incomplete calcination. Where unreacted carbonate leaves the kiln in the product or in collected dust, the calcination fraction F drops below 1.0. Glass melting is effectively complete (F = 1.0); some low-temperature ceramic firings are not. Apply a measured or documented F only — never assume incomplete calcination to reduce a figure without evidence.
- Non-carbonate alkali sources. Some modern glass batches substitute part of the soda ash with non-carbonate sources of sodium (e.g. sodium sulfate, or feldspar). These release no carbonate CO₂. Only the carbonate inputs enter the Tier 3 sum — counting a non-carbonate flux as soda ash overstates the process line.
- Recarbonation and absorbed CO₂. A minor fraction of process CO₂ can be reabsorbed by the product or by alkaline dust over time. IPCC Tier 1/2 ignores this; it is immaterial for glass and ceramics and is not netted off without a Tier 3 mass-balance and verifier agreement.
- Organic carbon in clay. Beyond carbonate decomposition, ceramic clays contain organic carbon that oxidises to CO₂ during firing. The ceramics loss factor (default 1.1) is the IPCC device for capturing this uplift; where organic content is measured, it can be modelled explicitly instead of via the multiplier.
- Soda ash vs limestone allocation. When a single batch sheet lists “carbonates” without splitting soda ash from limestone and dolomite, the Tier 3 sum cannot be built. Either obtain the split from procurement records or fall back to the Tier 1 output method — do not apply one carbonate’s factor to the whole undifferentiated mass.
EU ETS, MRV and CBAM
Glass and ceramic installations above the relevant capacity thresholds fall within the EU Emissions Trading System, which mandates a monitoring plan and annual verified reporting under the Monitoring and Reporting Regulation (MRR). The MRR’s calculation-based approach for process emissions is the same carbonate method set out here: activity data (carbonate or product mass) × an emission factor × where relevant a calcination/conversion factor.
Two MRV points specific to this sub-sector:
- Carbonate input vs. oxide output method. The MRR permits either Method A (based on carbonate inputs, equivalent to the Tier 3 inventory above) or Method B (based on the alkaline-oxide content of the product). Method A is the more common and maps directly onto the
carbonate.*stoichiometric factors. The chosen method is fixed in the monitoring plan and changed only with competent-authority approval. - Tier requirements scale with installation size. Larger emitters face higher minimum tier requirements and tighter uncertainty thresholds on activity data. A small ceramics kiln may use default factors; a major float-glass installation will be required to substantiate its carbonate inputs with weighing data.
For imported glass and ceramic goods, the EU Carbon Border Adjustment Mechanism requires embedded emissions to be reported. Where an importer cannot obtain verified producer-specific data, CBAM default emission factors apply — these are regulatory-reporting-tier values held separately in the MasterBrain CBAM family and are distinct from the IPCC process factors used for an installation’s own inventory. Do not substitute a CBAM default into an installation’s own Scope 1 calculation, or vice versa; they serve different purposes and carry different scope tags.
Uncertainty and Tier Guidance
The IPCC tier system is the language verifiers use to assess data quality. For glass and ceramics specifically:
| Tier | Method | Typical uncertainty | When acceptable |
|---|---|---|---|
| Tier 1 | Glass: product output × default EF. Ceramics: clay mass × default carbonate fraction. | ±30–60% | Screening; first-year inventory; immaterial sources |
| Tier 2 | Carbonate inputs × stoichiometric factors, country/site default composition | ±10–25% | Material sources; routine post-year-1 reporting |
| Tier 3 | Weighed carbonate inputs × stoichiometric factors × measured calcination fraction | ±2–7% | Large emitters; EU ETS high-tier obligation; reasonable-assurance audit |
Because glass batch composition is deliberately engineered and weighed, glass installations can reach Tier 3’s ±2–7% band relatively easily — the carbonate masses are already metered for quality control. Ceramics is harder: clay carbonate content varies within and between deposits, so reaching low uncertainty requires a sampling programme. Plan the data-quality progression for ceramics earlier, because the screening default carries the widest band of any figure in this methodology.
Verification and Assurance
Glass and ceramics process inventories face the standard assurance ladder, with sub-sector-specific evidence expectations:
- What auditors examine. Batch sheets and procurement records reconciling carbonate purchases to consumption; the carbonate/oxide method declared in the monitoring plan; the calcination fraction justification where F < 1.0; and — critically — evidence that the process and combustion lines are derived from independent activity data, not from one another.
- The disaggregation test. A verifier will check that process CO₂ and combustion CO₂ can each be traced to their own activity data and reconstructed independently. A merged figure, or a process line back-calculated from total fuel, fails this test regardless of whether the total is correct.
- Cullet substantiation (glass). The cullet fraction directly reduces the reported process figure, so verifiers substantiate it against cullet purchase and internal-recycling records. An unsupported cullet fraction is treated as zero.
The audit framework follows the GHG Protocol Corporate Standard for inventory boundaries, with the IPCC mineral-industry method supplying the calculation criteria and, for EU installations, the MRR and accredited-verifier regime supplying the assurance.
Error Traps with Magnitudes
Each error below produces a specific, quantifiable distortion. Magnitudes are shown for the FloatCo example (200,000 t glass, 30% cullet, 28,308 t process base) to make validation and disclosure risk concrete.
| Error | What happens | Magnitude (FloatCo) | How to avoid |
|---|---|---|---|
| Misclassify combustion fuel as process | Furnace fuel CO₂ added to the process line, doubling it and corrupting both benchmarks. | +55,900 t CO₂ phantom process line ~0.28 t/t × 200,000. Nearly triples the true 28,308 t process figure. |
Calculate process from carbonate mass, combustion from fuel quantity. Never derive one from the other. |
| Ignore the cullet credit | Tier 1 applied to total output without the (1 − fcullet) term. |
Overstates by 12,600 t CO₂ (+42.86%) 42,000 gross vs 29,400 net at 30% cullet. |
Always apply the cullet fraction to the Tier 1 output. In Tier 3 it is implicit (cullet adds no carbonate). |
Invent a minerals.ceramics.tier1_default key |
Calculation written against a non-existent output factor renders empty and silently zeros the ceramics source. | Entire ceramics line lost No Tier 1 ceramics factor exists; the figure must be built from clay × carbonate fraction. |
Use the carbonate-fraction method with minerals.ceramics.clay_* defaults and a real carbonate.* factor. |
| Wrong soda-ash key slug | Soda ash calculated with calcite’s factor (0.43971) instead of carbonate.sodium_carbonate (0.41492). |
+694 t CO₂ overstatement On 28,000 t soda ash: 28,000 × (0.43971 − 0.41492). |
Soda ash is carbonate.sodium_carbonate. There is no carbonate.soda_ash alias. |
| Use tier1_default where a product factor applies | Generic 0.20 used for float glass instead of the float-specific 0.21, or vice versa. | ~5% per-tonne mismatch 0.20 vs 0.21 = 0.01 t/t × output. Reverses sign for low-carbonate specialty glass. |
Match the slug to the prose claim — tier1_default is 0.20; the float/container rows are 0.21. |
| Assume incomplete calcination without evidence | F < 1.0 applied to glass melting to shave the figure, with no carry-over data. | Understatement = (1 − F) × total An unsupported F = 0.95 would understate by ~1,415 t CO₂. |
Use F = 1.0 for glass melting unless documented carbonate carry-over justifies otherwise. |
Methodology Metadata — for GHG Inventory Documentation
Copy into your GHG inventory methodology statement for inventory-transparency compliance. Adjust the tier and method lines to match what you actually applied.
Frequently Asked Questions
For the installation that operates the furnace or kiln, carbonate process emissions are Scope 1 — they occur at owned or controlled equipment. They sit on a separate Scope 1 line from the combustion emissions of the fuel used to heat the process, but both are Scope 1. For a company that buys finished glass or ceramic products, the producer’s process emissions appear in the buyer’s Scope 3 Category 1 (purchased goods), calculated from the embodied carbon of the product rather than from this methodology.
Cullet is glass that has already been melted once. Its carbonates decomposed during that first melt and released their CO₂ then. Remelting cullet rearranges already-formed glass — there is no fresh carbonate to decompose, so no new process CO₂. This is why the Tier 1 formula multiplies output by (1 − fcullet), and why cullet never appears in the Tier 3 carbonate inventory. Cullet does separately reduce combustion emissions by lowering the energy needed to melt the batch, but that is a distinct effect captured in the combustion line.
Use the carbonate-input (Tier 3) method wherever you weigh your batch inputs, which most glass installations do for quality control. It reflects your actual recipe and reaches ±2–7% uncertainty, and it is what reasonable-assurance audits and high-tier EU ETS obligations expect. The output-based Tier 1 method is best kept as a cross-check — if the two diverge by more than roughly 10%, investigate before reporting, because a large gap usually signals a batch-data error or a cullet-fraction problem.
No. The IPCC guidelines publish no per-tonne-of-product emission factor for fired ceramics, because carbonate content varies too widely between clay deposits to support one. Ceramics process CO₂ must be built from the clay’s carbonate fraction — clay mass × carbonate fraction × loss factor × the relevant carbonate stoichiometric factor. Where no site measurement exists, conservative default composition values stand in, but the figure is always computed, never looked up against an output factor.
No. Carbonate decomposition releases pure CO₂, whose global warming potential is 1 by definition under every IPCC assessment report. A tonne of process CO₂ is a tonne of CO₂e with no conversion. This differs from the combustion line, where small methane and nitrous oxide components require AR5 or AR6 GWP conversion — but those belong to the separate combustion inventory, not the carbonate process line covered here.
It is the same underlying chemistry — carbonate decomposition under heat — applied to a different industry. Cement and lime calcine limestone to make clinker and quicklime, and use per-tonne-of-product factors (clinker ~0.52, high-calcium lime ~0.75 t CO₂/t). Glass and ceramics calcine carbonates within a glass batch or clay body. The stoichiometric carbonate factors are shared across all of these; what differs is whether the calculation runs on product output, carbonate input, or carbonate fraction. The cement and lime calcination methodology covers the sibling case in full.