Is HVO Renewable Diesel Carbon-Neutral? Lifecycle Emissions by Feedstock, From Used Cooking Oil to Palm

1. Summary

No commercial hydrotreated vegetable oil (HVO) pathway is carbon-neutral today. The distinction that matters for buyers, investors and regulators is between two separate questions: whether a pathway emits less than fossil diesel over its full life cycle, and whether its net lifecycle emissions reach zero without external offsets. Every pathway examined here passes or fails the first test mainly according to land-use dynamics and the treatment of residues; none passes the second. Against the EU Renewable Energy Directive comparator of 94 g CO₂e/MJ, applied here as the single fossil baseline on a well-to-wheel, lower-heating-value basis [1], attributional central estimates place used cooking oil HVO at about 14 g CO₂e/MJ, an 85 percent saving [1][2][3]. Intermediate winter canola reaches about 32 g, a 66 percent saving [4], while main-season canola and rapeseed fall between about 51 and 70 g once central indirect land-use change (ILUC) values are added [1][3][4]. U.S. soybean HVO sits near 58 g on the same basis [3][5], palm oil HVO exceeds the fossil baseline at about 105 g once Indirect Land Use Change (ILUC) is counted [1][3], and palm grown on drained peat reaches roughly 320 g, several times the emissions of the diesel it displaces [6][7].

The report treats biogenic carbon as neutral, so the roughly 71 g of CO₂ released per megajoule of HVO burned is cancelled by the carbon the feedstock absorbed while growing, and by nothing more. Fossil hydrogen, which accounts for about 7 to 8 g/MJ and roughly half the footprint of waste-oil HVO [2], remains on the books, as do fertilizer manufacture and soil nitrous oxide, which reach about 18 g/MJ for rapeseed [1], together with process energy, ocean freight and land conversion. Depending on pathway, these residual burdens add between 12 and more than 300 g CO₂e/MJ. The gap between "lower than diesel" and "carbon-neutral" for mainstream pathways therefore runs between about 14 and 70 g CO₂e/MJ. Closing it physically would require renewable hydrogen, renewable process energy, verifiably surplus waste feedstock and durable carbon removal, yet by this report's stoichiometric estimate the removal available from capturing process CO₂ inside an HVO plant is at most about 4 g/MJ. Current neutrality claims consequently rest on accounting credits, soil-carbon claims or offsets rather than on the physical supply chain.

The Argonne National Laboratory study by Kwon, Liu, Cai and Wang illustrates the central mechanism. Renewable diesel from intermediate winter canola and from main-season spring canola shows nearly identical direct emissions, about 32 and 33 g CO₂e/MJ respectively, with farming contributing about 19 and 20 g/MJ; induced land-use change and related market effects, however, add about 18 g/MJ to the main-season crop and essentially nothing to the intermediate one [4]. The agronomic difference between the two systems is therefore about 1 g/MJ and the land-use difference about 18 g/MJ. The available evidence implies that this pattern generalizes: for the same crop and the same refinery, land dynamics decide the outcome more than any process variable.

For wastes and residues, attributional frameworks assign zero upstream burden, but consequential displacement analysis, which asks what replaces the material in its former use, changes the picture. The International Council on Clean Transportation (ICCT) estimates an additional 12.2 g CO₂e/MJ for U.S. used cooking oil HVO once diverted uses are replaced [8], and a 2025 ifeu study for Deutsche Umwelthilfe finds that when Asian-sourced used cooking oil is replaced by palm oil in its origin markets, the climate benefit can disappear entirely [2]. Traceability failures compound the problem: the European Commission's July 2025 review of suspected fraud in Chinese biodiesel imports identified weaknesses in certification audits without confirming fraud [9].

Three conclusions carry the most weight for decision-makers. The credible claim for the best HVO is a reduction of about 75 to 88 percent per megajoule, not neutrality. Crop-oil HVO carries a model-dependent penalty, because published ILUC values run from about 7 to more than 230 g CO₂e/MJ, a range wider than the entire direct footprint [10][11]. Finally, U.S. federal policy now diverges from European and Californian practice: the 2025 amendments to Section 45Z exclude ILUC for fuel produced after 2025 [12][13][14], whereas the EU retains its phase-out of high ILUC-risk fuels [15] and California caps the crediting of crop oils [16], so the same gallon carries materially different reported carbon intensities in different jurisdictions.

2. Scientific Background and System Boundary

2.1 HVO chemistry and its distinction from FAME

HVO, also termed hydroprocessed esters and fatty acids or renewable diesel, is produced by reacting triglycerides and free fatty acids with hydrogen over sulfided nickel-molybdenum or cobalt-molybdenum catalysts [17]. Oxygen leaves the molecule by one of two routes. Hydrodeoxygenation removes it as water and consumes the most hydrogen [18], whereas decarboxylation and decarbonylation remove it as CO₂ and CO, consuming less hydrogen at the cost of one carbon atom per chain and therefore lower yield [18][19]. The resulting straight-chain paraffins are isomerized, usually over noble-metal catalysts, to improve cold-flow properties [18], and the product slate comprises paraffinic diesel together with propane, renewable naphtha and light ends [17].

Fatty acid methyl ester (FAME) biodiesel differs in chemistry, co-products and use. It is made by transesterification with methanol, yields glycerol as a co-product and retains its ester oxygen [20], and fuel specifications restrict it to low blend levels, whereas HVO is a drop-in hydrocarbon usable neat [21][22][23]. These differences appear in the carbon accounts. For oilseed feedstocks, the Argonne analysis by Xu, Ou, Li, Hawkins and Wang found biodiesel to be the lower-emission product because transesterification is less energy-intensive than hydroprocessing; for high free-fatty-acid feedstocks such as tallow the ranking reverses, because FAME pretreatment becomes energy-intensive [20]. The ifeu inventory for used cooking oil shows the same trade-off at stage level: hydrogen contributes 7.8 g CO₂e/MJ and process energy 3.6 g/MJ to HVO, against 7 g/MJ for methanol and 7.5 g/MJ for process energy in FAME, the gap reflecting heat recovered from the exothermic hydrotreating reactions [2]. FAME serves below only as a comparison point.

2.2 System boundary, biogenic convention and baseline

The system boundary is cradle-to-grave, or well-to-wheel. It spans cultivation or collection; extraction or rendering and pretreatment; hydrogen supply; hydrotreating and isomerization; transport, distribution and combustion; direct and indirect land-use change; soil carbon where quantified; and residue displacement where consequential estimates exist. The embodied emissions of farm machinery and refinery infrastructure are excluded by the main regulatory methods [1], and they could not be quantified from HVO-specific primary sources (Section 4.5).

Biogenic carbon is treated as neutral, so both uptake and combustion CO₂ count as zero. As HVO is roughly 85 percent carbon by mass with a lower heating value of about 44 MJ/kg, combustion releases, by this report's calculation, about 0.85 × 3.67 / 44, or roughly 71 g of biogenic CO₂ per megajoule. Uptake offsets exactly that quantity and nothing more. Soil-carbon gains are counted only as separate stock changes, never as a second credit for carbon already embodied in the fuel, and carbon held in short-lived products is never treated as permanent removal.

The single fossil baseline is the 94 g CO₂e/MJ comparator set in Annex V of Directive (EU) 2018/2001 and retained under RED III, Directive (EU) 2023/2413 [1][24]. Other frameworks use different comparators: the 2009 Renewable Energy Directive used 83.8 g [25], CORSIA uses 89 g for fossil jet fuel [26] with an eligibility threshold near 80 g [27], and U.S. programs use petroleum diesel as modeled in GREET. Every savings percentage in this report is recomputed against 94 g; Argonne's reductions of 40 to 69 percent for oilseed pathways, for example, are measured against GREET's own baseline and are not carried over directly [20].

2.3 Global warming potential basis

The Annex V defaults use 100-year global warming potentials (GWPs) of 25 for methane and 298 for N₂O [1], which the IPCC's Sixth Assessment Report revises to 27 and 273, as ifeu notes [2]; Kwon et al. apply the AR6 100-year values [4]. The N₂O revision lowers any soil-N₂O term by 8.4 percent, which by this report's calculation amounts to about 1.5 g/MJ for RED rapeseed HVO, small relative to model uncertainty. The choice of metric matters more for methane-heavy palm oil from mills with open effluent ponds, where a 20-year basis would roughly triple the methane contribution. Figures are reported in each source's own GWP basis, and the resulting inconsistency is below 2 g/MJ for every non-palm pathway.

3. Key Stakeholders

Production is concentrated among a small number of firms. Neste (HEL:NESTE) operates HVO plants in Finland, the Netherlands and Singapore, and the ifeu study, citing the company, credits it with 69 percent of global capacity [2], a producer-sourced figure that predates the U.S. capacity build-out. Diamond Green Diesel, the joint venture of Valero Energy (NYSE:VLO) and Darling Ingredients (NYSE:DAR), pairs refining with North America's largest rendering network, and its California pathway documents apply CARB's default ILUC values of 29.1 g CO₂e/MJ for soybean oil and 14.5 g for canola [28][29]. Among converted refineries, the Rodeo facility of Phillips 66 (NYSE:PSX) lists soybean oil, canola, distillers corn oil, used cooking oil and animal fat in its CARB filings and states that it does not process palm [30]; Marathon Petroleum (NYSE:MPC) operates at Martinez; Eni (BIT:ENI) operates biorefineries at Venice and Gela [31]; and ifeu identifies TotalEnergies (EPA:TTE) as a European competitor [2].

Upstream, renderers, used cooking oil aggregators in China and Southeast Asia, and oilseed crushers control the lowest-carbon inputs. ISCC, the largest voluntary certification scheme recognized under the RED, sits at the point where waste status is attested and has faced criticism over audit weaknesses [32]. The methods themselves are set by the European Commission and its Joint Research Centre, the California Air Resources Board, the U.S. Environmental Protection Agency, the U.S. Treasury and Internal Revenue Service working with the Department of Energy's GREET team, and ICAO. On the demand side, heavy-road fleets, off-road and backup-power users and airlines compete for the same lipid feedstocks.

4. Technical and Operational Considerations

4.1 The canola benchmark: same crop, different land dynamics

Kwon et al. used Argonne's R&D GREET model to compare intermediate winter canola, grown in double- or relay-cropping between main crops, with main canola, predominantly spring canola occupying the field for a full season [4]. Direct lifecycle emissions of renewable diesel are about 32 g CO₂e/MJ for the intermediate system and 33 g for main spring canola, against about 30 and 31 g for biodiesel, with farming contributing about 19 and 20 g/MJ [4]. The inventory contains a counter-intuitive detail: because of the fertilizer product mix, upstream nitrogen emissions are about 5 g CO₂e per gram of nitrogen for intermediate canola against 3 g for spring canola, yet the intermediate crop still shows lower field N₂O, urea CO₂ and farming energy per pound of seed because of its higher yield [4]. On direct emissions, the two systems are effectively tied.

The decisive difference arises through markets. Induced land-use change and related effects add about 18 g CO₂e/MJ to main spring canola renewable diesel and 17 g to its biodiesel, and the authors conclude that the two systems may differ substantially because of their different land-use dynamics [4]. By this report's calculation from those figures, main-season canola HVO totals roughly 51 g CO₂e/MJ, a 46 percent saving against 94 g, while the intermediate crop totals roughly 32 g, a 66 percent saving. The general lesson, supported by the ILUC literature reviewed in Section 4.7, is that a feedstock grown on land that would otherwise lie fallow within an existing rotation does not displace food production, whereas one occupying a full season does. That conclusion depends on additionality, since the intermediate crop must not depress the yield of the main crop that follows, and Kwon et al. model this effect rather than observe it [4].

4.2 Cultivation inputs and soil nitrous oxide

Cultivation is the largest direct emission term for every purpose-grown pathway. The RED Annex V typical values put total cultivation emissions at 33.4 g CO₂e/MJ for rapeseed HVO, of which soil N₂O contributes 18 g; at 26.9 g for sunflower, with 12.5 g of soil N₂O; at 22.1 g for soybean, with 13.7 g; and at 27.3 g for palm, with 16.9 g [1]. For rapeseed, soil N₂O alone therefore represents 39 percent of the 45.8 g/MJ typical total [1]. Most of the remainder is fertilizer manufacture, dominated by ammonia synthesized from natural gas; lime, pesticides, seed and field diesel are smaller terms that the sources retrieved report only in aggregate.

N₂O is also the most uncertain term. The 2019 IPCC Refinement, as summarized by Hergoualc'h and colleagues and reported by Shorunke et al., replaced the single 1 percent direct emission factor with 1.6 percent for synthetic nitrogen in wet climates and 0.5 percent in dry climates [33]. Field measurements on winter oilseed rape generally point lower. Ruser et al. measured annual direct emissions of 0.24 to 5.48 kg N₂O-N per hectare at five German sites over three years and derived a rapeseed-specific factor of 0.6 percent at 200 kg N/ha [34]; a Danish field study found 0.51 percent [35]; and a German study measured indirect N₂O from ammonia volatilization up to 61 percent below the IPCC default and from leaching 64 to 89 percent below [36]. Evidence also runs the other way: a meta-analysis by Walter and colleagues found annual N₂O from winter oilseed rape about 22 percent higher than from winter cereals [33], and a 25-year process-model simulation of German rapeseed produced emission factors predominantly well below the aggregated IPCC default of 1 percent, but with strong variation between sites [37].

The implication is a range rather than a single correction. Treating the entire soil-N₂O term as fertilizer-driven, this report's illustrative calculation finds that scaling the RED rapeseed term from an effective 1 percent down to 0.6 percent removes roughly 5 to 7 g CO₂e/MJ, while scaling it up to 1.6 percent adds roughly 8 to 11 g. On N₂O assumptions alone, rapeseed HVO's direct total could therefore plausibly lie between about 39 and 57 g CO₂e/MJ. German field data support the lower half of that range, but they cannot be transferred to wetter, higher-input or tropical systems. For soybean, which fixes its own nitrogen, the N₂O term derives mostly from crop residues, and no field evidence specific to soybean HVO was retrieved.

4.3 Irrigation, machinery, drying, extraction and crushing

The principal HVO oilseeds, European rapeseed, North American canola and soybean, and palm, are mostly rainfed, and no inventory retrieved identifies irrigation as a separately material term; no credible pathway-specific source was identified for irrigated sunflower or soybean. Field fuel for tillage, planting, spraying and harvest is embedded in the GREET and RED farming totals, and Kwon et al. assume equal energy per acre for both canola systems [4]. The RED excludes emissions from machinery manufacture [1], no estimate specific to oilseed HVO was found, and drying and storage are folded into farming energy.

Oil extraction is a small term for temperate oilseeds, with RED typical values of 3.1 g CO₂e/MJ for rapeseed, 3 for sunflower and 3.3 for soybean [1][4]. CORSIA documentation notes that EU solvent extraction carries lower emissions because it is more efficient [38], but no source retrieved compares mechanical pressing with hexane extraction on a like-for-like HVO basis. Palm is the decisive exception: extraction contributes 3.8 g/MJ where the mill captures methane but 21.9 g/MJ where palm oil mill effluent decays anaerobically in open ponds [1][39]. Published Malaysian palm biodiesel values of 33.19 g CO₂e/MJ without biogas capture and 21.20 g with it confirm the scale of the difference [40][41]. Meal and cake are addressed through co-product allocation in Section 4.11.

4.4 Hydrogen supply

Hydrogen demand depends on feedstock chemistry and on which reactions dominate. Patent literature cites published consumption for deoxygenating and saturating soybean oil of about 250 to 320 Nm³ per cubic metre of oil [42][43], which this report converts to about 24 to 31 kg per tonne, or 2.4 to 3.1 percent by weight. A patent example for rapeseed oil reports make-up hydrogen of about 48 kg per tonne, a figure that includes losses [44]. Unsaturated oils such as soybean and sunflower require more hydrogen than saturated feeds such as palm, tallow and much used cooking oil, and Li et al. found that high-oleic soybean oil lowers renewable diesel emissions by about 1 g CO₂e/MJ through reduced hydrogen use [5]. Operating the process to favor decarboxylation saves hydrogen at the expense of yield [19][45]. These are patent examples and model inputs rather than audited plant data.

The carbon cost of that hydrogen depends almost entirely on its production route. ifeu assigns 7.8 g CO₂e/MJ to grey hydrogen, produced from natural gas without capture, at about 11 kg CO₂e per kilogram [2], which by this report's calculation implies about 0.71 g of hydrogen per megajoule of HVO. Applying a range of 0.71 to 0.85 g/MJ to cover saturated and unsaturated feeds, the report's calculations translate published hydrogen intensities into HVO burdens as follows. The IEA's Global Hydrogen Review 2024 gives 10 to 12 kg CO₂e/kg for steam methane reforming without capture [46], and its 2023 definitions report gives 10 to 14 kg, of which upstream methane and CO₂ contribute 1 to 5 kg [47]; this corresponds to about 7 to 12 g CO₂e per megajoule of HVO. Retrofit capture of about 60 percent of feedstock CO₂ yields 5 to 8 kg CO₂e/kg [46], or about 3.6 to 6.8 g/MJ, and capture above 90 percent yields 0.8 to 6 kg [46], or about 0.6 to 5.1 g/MJ, although the IEA notes that no plants using these high-capture technologies are yet in operation [46].

The width of the blue hydrogen range reflects methane leakage and realized capture. Bauer et al. show that with 93 percent plant-wide removal and 0.2 percent methane leakage, blue hydrogen cuts emissions by about 75 percent relative to grey on a 100-year basis, whereas at 8 percent leakage the reduction falls to about 45 percent; blue hydrogen reaches a competitive 2 to 3.5 kg CO₂e/kg only with high removal and leakage below about 1 percent, while operating first-of-a-kind plants such as Quest and Port Arthur remove only 50 to 60 percent [48]. Howarth and Jacobson find blue hydrogen only 9 to 12 percent below grey [49][50], though their assumptions of 3.5 percent leakage, a 20-year GWP and 65 percent capture are contested. Electrolysis on grid power beats reforming only where grid intensity falls below 200 to 240 g CO₂/kWh, according to the IEA [46]; at about 55 kWh per kilogram [48] and a 300 to 400 g/kWh grid, this report calculates about 17 to 22 kg CO₂e/kg, or about 12 to 19 g per megajoule of HVO, worse than grey hydrogen. Electrolysis on dedicated wind power produces about 1 kg CO₂e/kg [48], or about 0.7 to 0.9 g/MJ. Nuclear-powered electrolysis would likely yield similar results, but no primary source was identified for that figure. Reforming the plant's own propane and light ends releases biogenic CO₂, which counts as zero under the adopted convention, but it forfeits a co-product that would otherwise displace fossil LPG or naphtha, and no peer-reviewed quantification for commercial HVO was identified.

With grey hydrogen, hydrogen accounts for about half of used cooking oil HVO emissions, 7.8 of 15.8 g/MJ in the ifeu inventory [2], and, by this report's calculation, about 15 to 18 percent of RED rapeseed HVO direct emissions [1]. Hydrogen is therefore the largest physical lever for waste-oil HVO and a secondary lever for crop-oil HVO.

4.5 Hydrotreating, isomerization, catalysts, infrastructure and plant configuration

Heat and electricity for hydrotreating and isomerization contribute 3.6 g CO₂e/MJ in the ifeu inventory [2]. RED typical values for processing, excluding extraction, are about 7.4 to 7.8 g/MJ for vegetable-oil HVO including hydrogen, rising to 10.2 g/MJ for used cooking oil and 14.5 g/MJ for animal fats because processing absorbs pretreatment and, for fats, rendering at 4.3 g/MJ [1]. These are default values constructed by the Joint Research Centre rather than figures reported by plants [51][52]. Plant-reported data are rare. A 2026 Journal of Cleaner Production study drawing on primary data from Eni's Venice and Gela biorefineries reports 10.6 to 17.9 g CO₂e/MJ at Venice and 15.2 to 22.2 g at Gela across used cooking oil, palm oil mill effluent, animal fat and castor oil, and 20.2 to 21.9 g/MJ for product distributed in France, Italy and the Netherlands [31]. These inventories are peer-reviewed but not verified by a third party.

Hydrotreating catalysts based on nickel-molybdenum and cobalt-molybdenum, and platinum-group isomerization catalysts, are replaced on multi-year cycles, and spent material is usually reclaimed for its metals. No study retrieved quantifies their contribution per megajoule; the main inventories either omit them or fold them into minor chemicals, and the evidence suggests they fall below the noise, so no value is assigned. Embodied construction emissions are likewise excluded by the RED, which states that emissions from the manufacture of machinery and equipment are not taken into account [1]; GREET pathways also exclude them, ISO 14044 permits such cut-offs, and no HVO-specific amortized estimate was found.

Plant configuration matters mainly through hydrogen supply and energy integration. Garraín et al., as reported by Bonalumi et al., estimated about 12 g CO₂e/MJ well-to-tank for diesel co-processed with 8 to 13 percent soybean HVO, against 11.55 g for ultra-low-sulfur diesel, and about 24 g for pure soybean HVO [31]. A 2024 refinery-integration study found that co-feeding used cooking oil sharply raises a hydrotreater's hydrogen demand, while a new-build plant running entirely on used cooking oil could reach 93 percent savings [53], and coupling a waste-oil HVO process with high-temperature electrolysis has been modeled as a hydrogen source [54]. Evidence of systematic differences between standalone plants, converted refineries and co-processing remains thin, although converted refineries typically inherit grey-hydrogen reformers unless capture or electrolysis is added.

4.6 Global transport, distribution and combustion

RED typical values for transport and distribution are 1.7 g CO₂e/MJ for rapeseed HVO, 2 for sunflower, 7 for palm, 9.2 for soybean oil shipped from South America to Europe, 1.7 for used cooking oil and 1.5 for animal fats [1]. The waste-oil values assume a largely intra-regional supply chain that no longer matches actual trade [55]. ifeu models half of Germany's used cooking oil as collected in East Asia and shipped about 11,000 km, with the remainder trucked about 1,250 km within Europe and 450 km of product distribution, arriving at a total of 4.4 g/MJ, more than double the RED default, composed of 0.2 g for collection, 1.4 g for ocean freight, 1.9 g for road distribution and 0.8 g for service-station electricity [2]. Long-distance freight thus adds only a few grams per megajoule, but that is a large share of a small total, and it coincides with the locations where traceability risk is concentrated. At combustion, biogenic CO₂ counts as zero under the adopted convention; tailpipe methane and N₂O are minor in every framework reviewed, and no HVO-specific measurement showing them to be material was identified.

4.7 Land-use change, peat, soil carbon and carbon debt

Converting land to a fuel crop releases a one-off pulse of carbon, the carbon debt. Payback time in years equals that debt, allocated to the fuel, divided by the annual saving per hectare, where the annual saving is the fossil comparator minus the pathway's ongoing carbon intensity, multiplied by the fuel energy produced per hectare per year. Fargione et al. estimated that converting Southeast Asian lowland rainforest to palm releases about 702 t CO₂ per hectare with a payback of 86 years at 7.1 t per hectare per year saved; that converting peat rainforest to palm releases about 3,452 t with a payback of 423 years; that converting Amazon rainforest to soybean releases 737 t, with 39 percent allocated to biodiesel, and a payback of 319 years; and that converting Cerrado grassland to soybean carries a payback of 37 years [56]. This report ran an HVO-specific check assuming 3.7 t of palm oil per hectare, 0.84 t of HVO per tonne of oil, 44 MJ/kg, an ongoing intensity of 44 g/MJ and about 87 percent of the debt allocated to oil. Annual savings then come to about (94 − 44) × 136,000 MJ, or 6.8 t CO₂ per hectare, and the payback for a 611 t allocated debt to about 90 years.

Peat emissions recur each year rather than occurring once. Argonne's 2024 CCLUB update applies 38.1 t CO₂e per hectare per year for peat oxidation [6], and a figure cited by the ICCT is about 95 t CO₂ per hectare per year [7]. Spread over about 136 GJ of HVO per hectare, these add, by this report's calculation, about 280 and 700 g CO₂e/MJ respectively in every year of drainage, several times the roughly 50 g/MJ annual saving. Treated as a recurring flow, peat-grown palm HVO never repays its debt, which grows for as long as the land stays drained.

Soil carbon on existing cropland is a smaller term that can move in either direction. Reduced tillage and cover crops can raise soil organic carbon, but the gains are reversible, saturate over time and are difficult to verify. U.S. policy now allows farm-practice reductions to feed into tax credits through USDA's feedstock carbon-intensity framework, effective July 29th, 2026 [13][14].

Indirect land-use change is where the frameworks diverge most. An Argonne comparison published in 2020 places GTAP-BIO at the low end, at 29 g CO₂e/MJ, and GLOBIOM at the high end, at 231 g, for vegetable-oil pathways, with regulatory defaults clustered between 55 and 71.4 g [57]. The GLOBIOM analysis prepared for the European Commission by Valin et al., on a 20-year amortization, gives 231 g for palm, 150 g for soybean and 65 g for rapeseed; rescaled to CORSIA's 25 years by Taheripour and colleagues, those values become 185, 120 and 52 g [11], and by this report's arithmetic about 154, 100 and 43 g over 30 years. CARB's GTAP-BIO values are 71.4 g for palm, 29.1 g for soybean and 14.5 g for canola [58]; Argonne's CCLUB gives about 7.9 g for soybean biodiesel [10]; CORSIA's HEFA defaults are 24.5 g for U.S. and 27.0 g for Brazilian soybean and 24.1 g for EU and 26.0 g for global rapeseed [3]; and a newer GLOBIOM version gives 60.2 g for Malaysian and Indonesian palm, 50.4 g for U.S. soybean and 27.4 g for EU rapeseed [11]. CARB's soybean value is the mean of 30 model runs spanning yield-price elasticities, the degree to which yields respond to prices, of 0.05 to 0.35 [59].

The sources of divergence can be identified. CARB describes the yield-price elasticity as the most influential GTAP-BIO parameter, and its expert workgroup proposed handling double-cropping through region-specific elasticities of 0.175 to 0.3 [59]. Substitution between vegetable oils is treated very differently: EPA's 2010 analysis attributed only 3 percent of soybean-biodiesel land expansion to oil palm [60], whereas a 2025 GLOBIOM Monte Carlo study found palm consistently the primary substitute [61]. Land-cover emission factors also differ, and the ICCT notes that the Valin et al. and earlier MIRAGE studies used peat factors well below the value in the Page et al. review [60]. An empirical analysis by Chen, Sexton and Smith, discussed by the Union of Concerned Scientists, found deforestation largely independent of which vegetable oil is used for biofuel [58], a result that sits uneasily with feedstock-specific ILUC factors. All ILUC figures are model outputs rather than observations, a point a producer trade association stresses in its submission to CARB [62], but an effect being unobservable is not evidence that it is zero.

No regulatory framework charges fuel crops for the carbon the land would absorb if restored, although CORSIA's documentation treats foregone sequestration as a sensitivity case [38]. As an illustration, this report calculates from a yield of 3.3 t of seed per hectare and 42 percent oil content that rapeseed HVO delivers roughly 50 GJ per hectare and saves about 2.4 t CO₂ per hectare per year before allocation. Natural regrowth absorbing a few tonnes of CO₂ per hectare per year would match or exceed that saving, but no primary source was identified for the regrowth rate, so the comparison is only indicative.

4.8 Pathway results: purpose-grown oil crops

All savings below are measured against 94 g CO₂e/MJ [1]. Main-season European rapeseed HVO carries a RED typical value of 45.8 g CO₂e/MJ, with a default of 50.1, on an attributional, energy-allocated basis that excludes ILUC [1]. Adding CORSIA's EU rapeseed ILUC value of 24.1 g yields a central estimate of about 70 g, a 26 percent saving [3]. The range runs from 46 g, a 51 percent saving, to about 111 g under GLOBIOM's 20-year value of 65 g, 18 percent worse than fossil diesel [60], with the ILUC model and the soil-N₂O factor as the main drivers. The pathway is classified Tier 2, moving to Tier 1 under GLOBIOM.

U.S. main spring canola HVO combines about 33 g of direct emissions with about 18 g of induced effects, according to Kwon et al., for a central estimate of about 51 g and a 46 percent saving [4], within a range of about 33 to 65 g; it is classified Tier 2. Intermediate or double-cropped canola reaches about 32 g with no induced penalty, a 66 percent saving [4], within a range of about 28 to 45 g that depends on N₂O, yield and any penalty imposed on the following main crop; it is also Tier 2.

For soybean, Li et al. report 32.2 to 33.3 g CO₂e/MJ for U.S. renewable diesel under GREET mass allocation [5], while the RED typical value for soybean HVO delivered to Europe is 42.2 g, including 9.2 g of transport [1]. Adding a central ILUC of 25 g, bracketed by CORSIA's 24.5 g and CARB's 29.1 g, gives about 58 g for U.S.-consumed fuel, a 38 percent saving [3][63]. The range extends from 33 g with ILUC excluded, as under Section 45Z after 2025, to about 150 g with GLOBIOM's 25-year value of 120 g [11][14], spanning a 65 percent saving and a 60 percent increase. Soybean from newly converted Amazon or Cerrado land carries the carbon debt described in Section 4.7 and remains in Tier 1 for decades [56][64]; the central case is Tier 2.

Sunflower HVO carries a RED typical value of 39.4 g, with a default of 43.6, the lowest direct value among temperate crops [1], although HVO-specific evidence is the thinnest. The ICCT notes that sunflower biodiesel's modest savings turn into a result above fossil diesel once palm substitution is included [60]. With a central ILUC of about 30 g, compared with the RED Annex VIII provisional estimate of 55 g for oil crops [1], the central estimate is about 69 g, a 27 percent saving, within a range of 39 to 110 g. The pathway is Tier 2, bordering on Tier 1.

Palm oil on mineral soil, with no induced expansion, carries RED typical values of 44 g with methane capture at the mill, a 53 percent saving, and 62.1 g with open effluent ponds, a 34 percent saving with a default of 73.2 g [1]. A 2026 Journal of Cleaner Production study of 21 Chinese factories reported crude palm oil HVO at 87.19 to 94.14 g CO₂e/MJ, essentially at the fossil baseline [65]. Adding a central ILUC of about 60 g, between GLOBIOM's value for CORSIA of 60.2 g and CARB's 71.4 g [11][58], gives about 105 g with methane capture and about 122 g without it, 12 and 30 percent worse than fossil diesel respectively, and the range extends to about 275 g under Valin et al. [60]. The central case is Tier 1; Tier 2 holds only under the implausible assumption that additional palm demand causes no expansion. Palm on drained peat combines RED direct emissions of 44 to 62 g [1] with 280 to 700 g/MJ of recurring peat oxidation [6][7] for a central estimate of about 320 g, within a range of about 200 to 750 g, roughly 240 percent worse than fossil diesel at the centre. It is Tier 1, with emissions compounding in every year of drainage.

4.9 Pathway results: intermediate, cover-crop and marginal-land oilseeds

Carinata, camelina and pennycress are promoted as fitting within existing rotations or on marginal land, and regulators have largely accepted that framing. A CARB pathway report records a zero ILUC factor for camelina [66], Washington's Department of Ecology adopted zero ILUC for cover crops consistent with California and Oregon [67], CORSIA derives a default core value of 42 g CO₂e/MJ for camelina HEFA while noting higher nitrogen and phosphorus inputs for Canadian camelina [68], and Xu et al. included carinata among oilseed pathways achieving 40 to 69 percent reductions against GREET diesel [20]. The evidence supports a central estimate of about 35 g CO₂e/MJ, a 63 percent saving, within a range of about 25 to 50 g driven by yield, which is typically well below canola, by nitrogen and N₂O, and by any displacement of the main crop. These pathways are classified Tier 2.

Their low land-use burden rests on modeled additionality, and it weakens if intermediate crops delay planting and reduce main-crop yields, are grown as sole crops on food land, or simply substitute for other oils within an interchangeable global market. No market-mediated evidence at commercial scale was identified.

4.10 Pathway results: wastes, residues and by-products

Attributional values for used cooking oil HVO cluster tightly: the RED typical value is 11.9 g CO₂e/MJ, with a default of 16.0 [1]; CORSIA gives 13.9 g [3]; ifeu's inventory for Asian imports gives 15.8 g [2]; a Chinese factory study reports 6.54 to 10.18 g [65]; and Xu et al. report 12 to 19 g across waste feedstocks, with collection and rendering about 35 percent of well-to-wheel emissions [20]. The central attributional estimate is about 14 g, an 85 percent saving, within a range of about 7 to 19 g driven by hydrogen source and transport distance. As cited by ifeu, Germany's Federal Office for Agriculture and Food reported average savings of 87.16 percent for waste-based HVO in 2023 [2].

Consequential results diverge sharply according to origin. The ICCT found that in 2017 U.S. used cooking oil went to livestock feed, 50 kt, and oleochemicals, 1 kt; assuming half of new demand is met by diverting those uses, with soybean and palm oil as replacements, it estimates 12.2 g CO₂e/MJ of indirect emissions within a range of 2 to 13.2 g [8], which raises the pathway to about 26 g and a 72 percent saving. ifeu takes a more pessimistic view of Asian supply: its palm-oil penalty, including estimated land-use change, is 87.5 g CO₂e per megajoule of displaced used cooking oil, exceeding the 75.6 g net saving it calculates for the pathway [2]. This report infers from ifeu's components a consequential intensity on the order of 105 g per megajoule of HVO, which would place the pathway in Tier 1.

Supply limits bind. ifeu records that 89 percent of the used cooking oil behind German biofuels is imported, more than half of it from other continents [2]. Transport & Environment, citing Stratas Advisors in work commissioned by an advocacy group, estimated that Europe used close to 7 million tonnes for biofuels in 2023, close to eight times estimated collected volumes [69]. The Irish Examiner reported that Indonesia put its residue capacity at 300,000 tonnes while exporting almost five million tonnes in 2023 [70], a secondary press report for which no primary Indonesian data were identified. In its July 2025 conclusion the European Commission found audit weaknesses without confirming fraud and responded with tighter certification rules, stronger oversight of voluntary schemes and full roll-out of the Union Database, with key reforms expected by 2026 [9]. Enforcement has so far shown that it can detect weaknesses in the paperwork chain; it has not shown that it can verify waste status at current trade volumes. In Europe the counterfactual fate of this oil is discharge to municipal wastewater [2], which justifies collection but not an avoided-emissions credit, and no framework grants one. Used cooking oil HVO is therefore Tier 2 on attributional accounting and between Tier 2 and Tier 1 on consequential accounting, depending on origin.

For animal fats, the RED typical value for HVO from rendered fats is 16 g CO₂e/MJ, with a default of 21.8, applying only to Category 1 and 2 material under Regulation (EC) No 1069/2009 and excluding emissions from hygienisation [1]; CORSIA gives 22.5 g for tallow [3]. The central attributional estimate is about 19 g, an 80 percent saving, within 16 to 23 g, placing the pathway in Tier 2. Category matters for displacement: Category 1 and 2 materials face restrictions on feed use and go mainly to energy recovery, whereas Category 3 fats compete in feed, pet food and oleochemicals, and Cerulogy's 2023 report for Transport & Environment identifies palm oil as their most likely substitute [71]. U.S. oleochemical producers told C&EN that biofuel policy had raised tallow prices from 13 to 16 cents per pound to 39 cents, against 47 cents for soybean oil, and that supply is fixed by livestock production and does not respond to price [72]. Across ten U.S. waste and residue cases, the ICCT found indirect emissions ranging from −49 to +176 g CO₂e/MJ [8]; this report infers, without a published estimate to support it, that fully displaced Category 3 tallow replaced by palm could approach the fossil baseline.

Distillers corn oil HVO carries a CORSIA value of 17.2 g CO₂e/MJ [3]; Xu et al. identify it as the lowest-emission waste pathway because it bears none of the ethanol plant's emissions [20], and CARB assigns it no ILUC [58]. The central attributional estimate is about 17 g, an 82 percent saving, within about 12 to 20 g. Because the oil has an established market in animal feed, consequential values would be higher, but no source quantifying them was retrieved; the pathway is Tier 2 on attributional accounting.

Palm fatty acid distillate carries a CORSIA value of 20.7 g, a 78 percent saving [3]. It is a co-product of palm refining with established markets in feed, including rumen-protected fats, soap, oleochemicals and, historically, boiler fuel, and Cerulogy found the current split between those uses undocumented [73]. Treating it as burden-free is weakly justified: an Argonne analysis found that when it is treated as a co-product sharing palm's induced land-use change, it may deliver no greenhouse-gas reduction [57], and a consequential assessment would bring it close to palm oil itself. It is Tier 2 on attributional accounting, with Tier 1 plausible on consequential accounting. Crude tall oil has an established pine-chemicals market, but no verifiable HVO-specific intensity was retrieved, so no central estimate is given; it is provisionally Tier 2 with material displacement risk. Oil recovered from palm oil mill effluent is in principle a genuine low-value residue but carries the most acute integrity concerns, illustrated by an unverified industry claim, reported by S&P Global, that 1.8 million tonnes of fraudulent ISCC-certified material entered the EU in 2023 [32].

4.11 Accounting methodology and co-product allocation

The RED, CORSIA core values and GREET direct results are attributional, dividing up the emissions of existing supply chains, whereas ILUC factors and displacement analyses are consequential, estimating the change caused by additional demand. The choice between them reverses the verdict in three cases: palm oil moves from a 53 percent saving to worse than fossil diesel; Asian-sourced used cooking oil moves from an 85 percent saving to potentially worse than fossil diesel; and soybean under GLOBIOM moves from a 65 percent saving to potentially worse than fossil diesel. For intermediate oilseeds, and for European used cooking oil drawn from genuinely new collection, the two approaches largely agree, which makes these the most robust claims in the dataset.

Co-product allocation also moves results. The RED allocates by energy content with no further co-product credit [1], while GREET's default for soybean is mass allocation between oil and meal [5][74]. Soybean oil represents roughly 19 percent of crushed mass but about a third of its energy content, so energy allocation assigns the oil, by this report's calculation, roughly 1.8 times the upstream burden that mass allocation does, enough to shift the cultivation term by more than 10 g CO₂e/MJ. Economic allocation lands close to energy allocation for soybean. System expansion, which credits the meal for the protein feed it displaces, can cut the oil's burden sharply or even make it negative, depending on which product is displaced and on that product's ILUC. HVO's propane, naphtha and light ends are allocated by energy in the RED [1], whereas under system expansion they would earn credits for displacing fossil LPG and naphtha. Xu et al. identify land-use change and allocation, together with fertilizer, N₂O and conversion inputs, as the key drivers of results [20].

ISO 14044 prefers avoiding allocation where possible, by subdividing processes or expanding the system, then allocating by physical relationships, and only then by economic value, while ISO 14067 requires transparent reporting of biogenic carbon. No regulatory framework follows the ISO hierarchy strictly.

4.12 Results per tonne-kilometre and per kilowatt-hour

For heavy road freight, this report assumes a long-haul 40-tonne tractor-trailer consuming about 32 litres per 100 km, or about 11.5 MJ/km, with a 19-tonne average payload, which gives about 0.6 MJ per tonne-kilometre. Engine efficiency is taken as equal for HVO and diesel; HVO's lower volumetric energy density, about 34 against 36 MJ per litre in the RED [1], raises litres consumed by a few percent without materially changing energy per kilometre. These are the report's own assumptions rather than sourced fleet data. For stationary generation, the report assumes a diesel generator at 38 percent electrical efficiency on a lower-heating-value basis, requiring about 9.5 MJ of fuel per kilowatt-hour.

On those assumptions, and by this report's calculation from the pathway estimates above, fossil diesel emits about 56 g CO₂e per tonne-kilometre and about 890 g per kilowatt-hour. Used cooking oil HVO emits about 8 g per tonne-kilometre, or 16 g on consequential accounting, and about 130 g per kilowatt-hour; tallow HVO about 11 g per tonne-kilometre; intermediate canola HVO about 19 g per tonne-kilometre and about 300 g per kilowatt-hour; soybean HVO about 35 g per tonne-kilometre; EU rapeseed HVO about 42 g per tonne-kilometre and about 660 g per kilowatt-hour; mineral-soil palm with methane capture and ILUC about 63 g per tonne-kilometre and about 1,000 g per kilowatt-hour; and peat palm about 190 g per tonne-kilometre and about 3,000 g per kilowatt-hour. For comparison, a 2025 study in Fuel reports results from Chatterjee et al. of 380 g CO₂e/kWh for palm-based HVO and 700 g for rapeseed-based HVO [17]. The RED comparator for liquid biofuels used in electricity is 183 g CO₂e per megajoule of electricity, about 660 g/kWh [1], so generator operators reporting against it would see smaller percentage savings than transport users.

4.13 Carbon-neutrality classification

The classification applies four tiers defined against 94 g CO₂e/MJ [1]. Tier 1 covers pathways above 94 g, emitting more than fossil diesel. Tier 2 covers pathways above 5 g and up to 94 g, lower than fossil diesel but net-positive. Tier 3 covers pathways above 0 g and up to 5 g, which the report treats as near-zero, corresponding to savings of about 95 percent or more. Tier 4 covers pathways at or below zero, which are net-zero or net-negative.

On central estimates, palm on drained peat, palm with or without methane capture once ILUC is included, and soybean from newly converted land fall in Tier 1. Main-season rapeseed and canola, U.S. soybean, sunflower and intermediate oilseeds fall in Tier 2, although rapeseed, soybean and sunflower face Tier 1 exposure under GLOBIOM-type ILUC values. Used cooking oil, tallow, distillers corn oil and palm fatty acid distillate are Tier 2 on attributional accounting, while Asian-sourced used cooking oil, Category 3 tallow and palm fatty acid distillate move toward or into Tier 1 on consequential accounting. No pathway reaches Tier 3 or Tier 4 on central estimates.

Tier 3 is physically plausible only for waste-oil HVO, and only if renewable electrolytic hydrogen replaces grey hydrogen, removing about 7 g/MJ and leaving 0.7 to 0.9 g; if renewable heat and power eliminate most of the 3.6 g/MJ energy term; if local collection and electrified distribution cut transport from about 4.4 g toward 1 to 2 g; and if the feedstock is verifiably surplus waste with no displacement. Under those conditions, used cooking oil HVO would fall to roughly 3 to 5 g CO₂e/MJ. That figure is this report's forward projection from ifeu's stage-by-stage breakdown [2], not a demonstrated result, and no commercial supply chain is documented as meeting all these conditions today. Crop-oil pathways cannot physically reach Tier 3, because fertilizer and soil N₂O alone exceed 5 g/MJ [1].

Tier 4 requires durable carbon removal, and the removal potential inside an HVO plant is small. Even a process that removed all feedstock oxygen by decarboxylation would release at most about 150 kg of biogenic CO₂ per tonne of oil, which this report's stoichiometric estimate converts to about 4 g CO₂ per megajoule of HVO; processes dominated by hydrodeoxygenation release much less. Capturing CO₂ at a fossil reformer lowers hydrogen emissions but creates no negative flow. Claims of near-zero or negative intensity therefore depend on soil-carbon credits of uncertain permanence, which are now eligible under Section 45Z through USDA's framework [14], on credits for displaced co-products or system expansion, on avoided-emissions credits that no framework grants for used cooking oil, on the exclusion of ILUC, or on external offsets. Such claims achieve neutrality through accounting rather than through the physical supply chain.

5. Economic and Market Dynamics

Incentives based on carbon intensity have turned a feedstock's carbon intensity into its price. As reported by the Irish Examiner, EU demand pushed used cooking oil to about €1,400 per tonne by February 2022, almost double its February 2020 price and above the price of virgin vegetable oil [70]. A waste priced above the product from which it derives signals scarcity and creates an incentive for fraud. The market now prices feedstock origin explicitly: Fastmarkets publishes separate HVO price assessments for crop, used cooking oil, Category 3 tallow and palm oil mill effluent pathways [21].

Headroom in U.S. supply appears limited. A January 2022 Cerulogy analysis by Malins and Sandford, supported by the ICCT, found from OECD-FAO data that 14 percent of U.S. virgin vegetable-oil consumption already went to biofuels. It estimated that growth in domestic soybean-oil production could support perhaps another 300 million gallons of fuel production, and increased use of waste and residual oils another 150 million, against more than 5 billion gallons a year of renewable diesel capacity announced for 2024; beyond that, supply would require dramatic increases in canola and palm oil imports or large-scale displacement of other uses [90]. Virgin oils already matter in California: the American Soybean Association, citing CARB data, reports that they supplied 175 million gallons of biomass-based diesel in the first quarter of 2024, about 30 percent of volumes shipped into the state [75].

Competition between road HVO and sustainable aviation fuel (SAF) for the same lipids is structural, because HEFA, the process that produces HVO, dominates near-term SAF supply. As an analysis by studio Gear Up for European waste-oil stakeholders notes, aviation incentives are designed separately and could create unequal competition for limited feedstock [55]. Because residue supply does not grow with price, shifting it between sectors does not reduce total emissions; it moves the lowest-carbon lipids to the highest bidder and pushes the other sector toward crop oils. No recent primary price series was available in the sources retrieved, but the direction is clear from the evidence above: HVO demand beyond genuine residue supply is met at the margin by vegetable oil, and ultimately by the cheapest vegetable oil, palm.

HEFA Process: SAF Production & Molecular Sieve Purification

Master the HEFA Process for Sustainable Aviation Fuel (SAF) production. Learn how critical molecular sieve purification protects catalysts and boosts SAF yields to 75-80%.

Jalon

6. Regulatory Landscape

RED III retains the 94 g CO₂e/MJ comparator, attributional energy allocation and the Annex V typical and default values [1][24]. It excludes ILUC from compliance savings, reporting it instead through Annex VIII provisional estimates that include 55 g for oil crops [1], and it caps food and feed crop biofuels at the lower of 7 percent or each member state's 2020 share plus 1 percentage point [24]. Annex IX Part B feedstocks, principally used cooking oil and animal fats, are capped at 1.7 percent of transport energy, a limit Germany has raised to 1.9 percent [76][77]. High ILUC-risk biofuels are to be phased out by 2030, and the 2019 delegated regulation limited that category to palm oil [78]. Germany has not counted palm-based fuels toward its greenhouse-gas quota since 2023 [2].

The California Low Carbon Fuel Standard uses CA-GREET with GTAP-BIO ILUC values of 29.1 g for soybean, 14.5 g for canola, 71.4 g for palm and zero for wastes [28][29][58]. Its 2024 amendments, which took effect on 1 July 2025 [79], cap credit generation for biomass-based diesel from soybean, canola and sunflower oil at 20 percent of each company's annual volume, assigning any excess the diesel benchmark intensity [11][79][80]; bar palm-derived fuels from earning credits [81]; and require tracking to the point of origin, third-party certification and land already cultivated before 2008 [11][82]. California is the only major jurisdiction that converts ILUC concern into a volume limit.

The U.S. Renewable Fuel Standard relies on EPA's 2010 lifecycle analysis and later pathway rules, which include ILUC, and its 2018 grain sorghum oil pathway incorporated displacement emissions, which the ICCT identifies as the first U.S. regulatory use of displacement methodology [8]. The Section 45Z clean fuel production credit, as amended in 2025 by the OBBBA, requires emissions rates from DOE's 45ZCF-GREET model [83], excludes ILUC for fuel produced after 31 December 2025 [12][13], and restricts feedstocks to the United States, Mexico and Canada, making other imported used cooking oil ineligible after 2025 under IRS Notice 2026-53 [13][84]. Users of model versions released before June 2026 must subtract the ILUC value and apply a 1.055 conversion factor [13]. A commercial analysis by Crux reports that ILUC had contributed about 13 to 14 g CO₂e/MJ to soybean renewable diesel and more than 16 g to canola pathways in 45ZCF-GREET, against a credit threshold of about 47.4 g/MJ [85]. Excluding ILUC is a policy decision to omit a modeled emission, not a finding that the emission is zero.

ICAO's CORSIA combines a core lifecycle value with an ILUC value informed by GLOBIOM and GTAP, amortized over 25 years, against an 89 g comparator, and offers both default values and actual-value calculations [3][38]. ISO 14040/14044 and ISO 14067 are method standards rather than compliance schemes; they permit either attributional or consequential framing, set the allocation hierarchy described in Section 4.11 and require transparent reporting of biogenic carbon and direct land-use change.

7. Geopolitical and Strategic Dimensions

Feedstock supply is geographically concentrated. Palm oil and its residue streams come overwhelmingly from Indonesia and Malaysia, where expansion has historically involved clearing rainforest and draining peat [56], and ifeu records that almost half of the used cooking oil behind German biofuels in 2023 came from Asia, with China, Indonesia and Malaysia the main exporters [2]. Trade and fraud concerns have followed. The European Commission opened an anti-dumping investigation into Chinese biodiesel in December 2023, as reported by Transport & Environment [69]. France, Germany and the Netherlands urged EU energy ministers in May 2024 to tighten checks on biofuel production sites wherever they are located and to reject sustainability certification where auditors are refused access, amid industry concern that imports declared as recycled oil and fat were made from cheaper virgin oil [86]. In January 2025 Indonesia made exports of used cooking oil, palm oil mill effluent and palm residue subject to government allocation, citing feedstock needs for its B40 biodiesel mandate [87]; the U.S. Department of Agriculture's Jakarta office characterized the curbs as also discouraging the mixing of crude palm oil into residue exports [88]. The anti-dumping investigation concluded in February 2025 with definitive duties on Chinese biodiesel, HVO included, of 10 percent for one producer and 21.7 to 35.5 percent for other exporters, while sustainable aviation fuel was excluded [89]. The United States has taken a different route by making North American sourcing a condition of 45Z (Clean Fuel Production Credit) [84], which reduces exposure to Asian traceability risk but also removes ILUC from the accounting. As vegetable oils are interchangeable, biofuel demand anywhere draws on the global oil balance and on palm expansion at the margin, whatever origin is claimed at the refinery gate.

8. Risk Matrix

9. Strategic Recommendations

9.1 For fleet operators and power-generation buyers

Buyers should procure HVO on verified carbon intensity rather than on "plant-based" or "renewable" labels, and should report claims as percentage reductions against a stated comparator, never as carbon neutrality. On attributional accounting against 94 g CO₂e/MJ, a defensible claim for waste-oil HVO is a reduction of about 80 to 88 percent per megajoule, while for crop-oil HVO with ILUC it is closer to 25 to 50 percent. Contracts should require feedstock origin down to country and collection region, Union Database or equivalent traceability and the exclusion of palm and its derivatives, and conservative buyers should also request consequential intensity including displacement. Any net-zero operations claim must explicitly cover the residual 12 to 15 g/MJ or more with disclosed durable removals. For stationary power, where the best HVO comes to about 130 g CO₂e/kWh, the relevant comparison is the local grid combined with battery storage rather than diesel.

9.2 For producers, investors and corporate strategists

Decarbonizing hydrogen is the highest-return physical lever for waste-oil plants, roughly halving reported intensity, and investment cases should specify the hydrogen route, the capture rate and upstream methane leakage, tested against the IEA and Bauer et al. ranges [46][48] rather than nameplate capture rates. Crop-oil capacity should be valued under at least two ILUC regimes, the current 45Z exclusion and a GLOBIOM-type penalty, because the spread between them exceeds the direct footprint. Intermediate oilseeds are the most robust crop option, but their zero-ILUC treatment depends on main-crop yields being unaffected, so investors should fund field verification rather than assume it. Business plans should not assume that residue supply grows with demand.

9.3 For regulators and standard-setters

Regulators should publish carbon intensities that include ILUC and displacement alongside compliance values, update GWPs to AR6 with 20-year sensitivity cases for methane, and revisit the animal-fat default given that it applies only to certain categories. Traceability should be backed by physical verification rather than document audits alone. Carbon-neutrality marketing claims for HVO should be restricted to cases that demonstrate durable removals, a condition no commercial pathway currently meets.

10. Caveats

Several figures in this report are its own calculations from published components, identified as such in the text: the hydrogen burdens per megajoule, peat oxidation per megajoule, the carbon-debt check, the allocation ratio, the tonne-kilometre and kilowatt-hour conversions, the stoichiometric removal ceiling and the consequential used cooking oil figure inferred from ifeu. The full inventory tables of Xu et al. [20] and Kwon et al. [4] could not be accessed, so their stage-level results derive from abstracts and figure summaries. Most regulatory values are defaults constructed by the Joint Research Centre or from GREET rather than inventories measured at plants, and plant studies affiliated with producers are peer-reviewed but not verified by a third party. Evidence is thinnest for sunflower HVO, tall oil, catalysts, infrastructure and irrigation, and for the consequential values of tallow and distillers corn oil. Several market figures come from secondary press or advocacy sources and are flagged where used.

Inert-Anode Chemistry and the ELYSIS Smelter in Québec: Can Alcoa and Rio Tinto Deliver Carbon-Free Aluminum by 2027

ELYSIS proved inert-anode smelting at 450 kA in 2025. Whether Rio Tinto’s 2027 Arvida plant follows is the harder question.

DataDeep TechJohn D

Wind Turbine Lifecycle Emissions and Carbon Payback: Why 11 g CO2-eq/kWh Depends on Capacity Factor, Allocation, and the Displaced Grid

Utility-scale wind converges near 11 g CO2-eq/kWh, but capacity factor, allocation convention, and the displaced grid decide the carbon payback.

DataDeep TechJohn D

Can You Make Bioplastics in a Home Workshop? Eight Polymer Families, Numeric Recipes, Under $600 of Tooling

Eight bio-polymer families, 24 numeric recipes, under $600 of tooling. Drying energy costs $3 to $6 per kg, usually more than the polymer itself.

DataDeep TechJohn D

References

[1] European Parliament and Council of the European Union. 2018. "Directive (EU) 2018/2001 of the European Parliament and of the Council of 11 December 2018 on the Promotion of the Use of Energy from Renewable Sources (Recast), Annex V: Rules for Calculating the Greenhouse Gas Impact of Biofuels, Bioliquids and Their Fossil Fuel Comparators." Official Journal of the European Union L 328. Consolidated text, 7 June 2022. https://www.seai.ie/sites/default/files/publications/RED-II-directive-ANNEX-V-RULES-FOR-CALCULATING-THE-GREENHOUSE-GAS-IMPACT-OF-BIOFUELS-BIOLIQUIDS-AND-THEIR-FOSSIL-FUEL-COMPARATORS.pdf.

[2] ifeu (Institut für Energie- und Umweltforschung Heidelberg). 2025. Carbon Footprint of Biodiesel and HVO Based on Used Cooking Oil (UCO). Study for Deutsche Umwelthilfe. https://www.duh.de/fileadmin/user_upload/download/Projektinformation/Naturschutz/Agrokraftstoffe/2025-08-07_DUH_ifeu_study_GHG_balance_UCO_english.pdf.

[3] International Civil Aviation Organization. 2022. CORSIA Default Life Cycle Emissions Values for CORSIA Eligible Fuels. June. https://www.icao.int/sites/default/files/environmental-protection/Documents/ICAO-document-06-Default-Life-Cycle-Emissions-June-2022.pdf.

[4] Kwon, Hoyoung, Xinyu Liu, Hao Cai, and Michael Wang. 2026. "Life Cycle Analysis of Growing Canola for Biofuel Production in the United States." ACS Sustainable Chemistry & Engineering 14 (24): 11209–19. https://doi.org/10.1021/acssuschemeng.6c03620.

[5] Li, Y., H. Xu, D. Northrup, and M. Wang. 2023. "Effects of Soybean Varieties on Life-Cycle Greenhouse Gas Emissions of Biodiesel and Renewable Diesel." Biofuels, Bioproducts and Biorefining. https://doi.org/10.1002/bbb.2462.

[6] Argonne National Laboratory. 2024. Expansion of Carbon Calculator for Land Use and Land Management Change from Biofuels Production (CCLUB). ANL/ESIA-24/22. https://publications.anl.gov/anlpubs/2025/05/193639.pdf.

[7] International Council on Clean Transportation. n.d. "ILUC Factors and Palm Oil Yields: Resolving an Apparent Contradiction." ICCT staff blog. https://theicct.org/blog/staff/iluc-palm-oil-yields-resolving-apparent-contradiction.

[8] O'Malley, Jane, Stephanie Searle, and Nikita Pavlenko. 2021. Indirect Emissions from Waste and Residue Feedstocks: 10 Case Studies from the United States. White paper. Washington, DC: International Council on Clean Transportation. https://theicct.org/wp-content/uploads/2021/12/indirect-emissions-waste-feedstocks-US-white-paper-v4.pdf.

[9] European Commission. 2025. "Commission Concludes Examination of Potential Chinese Biofuel Imports Fraud." Transition Pathways news. https://transition-pathways.europa.eu/news-articles/commission-concludes-examination-potential-chinese-biofuel-imports-fraud.

[10] Life Cycle Associates. 2023. GHG Emissions Reductions due to the RFS2: A 2022 Update. Report LCA.6145.238.2023, prepared for the Renewable Fuels Association. https://ethanolrfa.org/file/2424/GHG%20Emissions%20Reductions%20due%20to%20the%20RFS2%20-%20Feb%202023.pdf.

[11] Taheripour, Farzad, Steffen Mueller, and Hoyoung Kwon. 2021. "Response to 'How Robust Are Reductions in Modeled Estimates from GTAP-BIO of the Indirect Land Use Change Induced by Conventional Biofuels?'" Journal of Cleaner Production 310: 127431. https://doi.org/10.1016/j.jclepro.2021.127431.

[12] Baker Botts. 2026. "IRS Issues Proposed Regulations Regarding § 45Z Clean Fuel Production Tax Credit." Thought Leadership, February. https://www.bakerbotts.com/thought-leadership/publications/2026/february/irs-issues-proposed-regulations-regarding-45z-clean-fuel-production-tax-credit.

[13] Environment, Land & Resources. 2026. "IRS Issues 45Z Tax Credit Notice and DOE Updates 45ZCF-GREET Model." September. https://www.globalelr.com/2026/09/irs-issues-45z-tax-credit-notice-and-doe-updates-45zcf-greet-model/.

[14] Trinity Consultants. n.d. "Section 45Z Update: GREET Model and USDA Feedstock Guidance." https://trinityconsultants.com/resources/section-45z-update-greet-model-and-usda-feedstock-guidance/.

[15] DieselNet. n.d. "Fuels: EU: Renewable Transportation Fuel Policies." https://dieselnet.com/standards/eu/fuel_renewable.php.

[16] DieselNet. 2024. "California Adopts Major Revisions to Low Carbon Fuel Standard." November. https://dieselnet.com/news/2024/11carb.php.

[17] Ajeeb, Wagd, Diogo Melo Gomes, Rui Costa Neto, and Patrícia Baptista. 2025. "Life Cycle Analysis of Hydrotreated Vegetable Oils Production Based on Green Hydrogen and Used Cooking Oils." Fuel 390: 134749. https://doi.org/10.1016/j.fuel.2025.134749.

[18] "Optimization of Energy Consumption in a Hydrotreating Process for Green Diesel Production from Palm Oil." 2016. In Computer Aided Chemical Engineering, vol. 38. Amsterdam: Elsevier. https://www.sciencedirect.com/science/article/abs/pii/B9780444634283501302.

[19] DieselNet. n.d. "Renewable Hydrocarbon Fuels." https://dieselnet.com/tech/fuel_renewable-diesel.php.

[20] Xu, Hui, Longwen Ou, Yuan Li, Troy R. Hawkins, and Michael Wang. 2022. "Life Cycle Greenhouse Gas Emissions of Biodiesel and Renewable Diesel Production in the United States." Environmental Science & Technology 56 (12): 7512ff. https://doi.org/10.1021/acs.est.2c00289.

[21] Fastmarkets. n.d. "Fastmarkets Price Launch: What Is Hydrotreated Vegetable Oil (HVO)?" https://www.fastmarkets.com/insights/fastmarkets-price-launch-what-is-hydrotreated-vegetable-oil-hvo/.

[22] "The European Energy Policy for 2020–2030 RED II: What Future for Vegetable Oil as a Source of Bioenergy?" 2019. OCL: Oilseeds and Fats, Crops and Lipids. https://www.ocl-journal.org/articles/ocl/full_html/2019/01/ocl190040s/ocl190040s.html.

[23] ETIP Bioenergy. 2020. "Hydrogenated Vegetable Oil (HVO)." Bioenergy Fact Sheet. https://old.etipbioenergy.eu/images/ETIP_B_Factsheet_HVO_feb2020.pdf.

[24] International Council on Clean Transportation. 2023. [Policy update on transport-fuel provisions in the European Union's revised Renewable Energy Directive (RED III)]. July. https://theicct.org/wp-content/uploads/2023/07/fuels-fit-for-55-red-iii-jul23.pdf.

[25] U.S. Patent 11,603,502. n.d. "Marine Fuel Compositions." https://image-ppubs.uspto.gov/dirsearch-public/print/downloadPdf/11603502.

[26] S&P Global Commodity Insights. 2024. Specifications Guide: Global Biofuels. December. https://www.spglobal.com/commodityinsights/PlattsContent/_assets/_files/en/our-methodology/methodology-specifications/global_biofuels.pdf.

[27] Sesame Sustainability. n.d. "Uncovering What Drives SAF Carbon Intensity: CORSIA LCA Analysis by Sesame." https://sesamesustainability.com/uncovering-what-drives-saf-carbon-intensity-corsia-lca-analysis-by-sesame/.

[28] California Air Resources Board. 2017. CARB LCFS Tier 2 Fuel Pathway Report: Soybean Oil Renewable Diesel (Diamond Green Diesel). https://ww2.arb.ca.gov/sites/default/files/classic/fuels/lcfs/fuelpathways/comments/tier2/dgd-soy-rd-rpt-030917.pdf.

[29] California Air Resources Board. n.d. LCFS Tier 2 Fuel Pathway Report B0492. https://ww2.arb.ca.gov/sites/default/files/classic/fuels/lcfs/fuelpathways/comments/tier2/b0492_report.pdf.

[30] California Air Resources Board. n.d. "California LCFS: B0825 Fuel Pathway Application Response to Comments." https://ww2.arb.ca.gov/sites/default/files/classic/fuels/lcfs/fuelpathways/comments/tier2/b0825_response.pdf.

[31] Bonalumi, Davide, Mehrshad Kolahchian Tabrizi, Jacopo Famiglietti, Letizia Bua, and Giammarco Gioco. 2026. "The Carbon Footprint of Hydrotreated Vegetable Oil (HVO) Production and Consumption Based on Primary Data." Journal of Cleaner Production 558: 148058. https://doi.org/10.1016/j.jclepro.2026.148058.

[32] Advanced BioFuels USA. n.d. "EU Scrutinizes Fraud in Certification of Biofuels." https://advancedbiofuelsusa.info/eu-scrutinizes-fraud-in-certification-of-biofuels.

[33] Shorunke, Akeem T., Bobbi L. Helgason, and Richard E. Farrell. 2025. "Evidence of the Need for Crop-Specific N2O Emission Factors." Soil Biology and Biochemistry 202: 109694. https://doi.org/10.1016/j.soilbio.2024.109694.

[34] Ruser, R., et al. 2017. "Nitrous Oxide Emissions from Winter Oilseed Rape Cultivation." Agriculture, Ecosystems & Environment 249. https://www.sciencedirect.com/science/article/pii/S0167880917303377.

[35] "Nitrous Oxide Emissions from Oilseed Rape Cultivation Were Unaffected by Flash Pyrolysis Biochar of Different Type, Rate and Field Ageing." 2020. Science of the Total Environment. https://www.sciencedirect.com/science/article/abs/pii/S0048969720316533.

[36] "Indirect Nitrous Oxide Emissions from Oilseed Rape Cropping Systems by NH3 Volatilization and Nitrate Leaching as Affected by Nitrogen Source, N Rate and Site Conditions." 2020. European Journal of Agronomy. https://www.sciencedirect.com/science/article/abs/pii/S1161030120300472.

[37] Räbiger, Thomas, Dorothee Neukam, Astrid Knieß, Ulf Böttcher, Henning Kage, and Insa Kühling. 2024. "Long-Term Simulated Direct N2O Emissions from German Oilseed Rape Cultivation below the IPCC Emission Factor." Agriculture 14 (1): 70. https://doi.org/10.3390/agriculture14010070.

[38] International Civil Aviation Organization. 2019. CORSIA Supporting Document: CORSIA Eligible Fuels, Life Cycle Assessment Methodology. https://cdn.prod.website-files.com/643691764f0ee331841022ac/643691764f0ee302271027ad_2019_ICAO_CORSIA%20Supporting%20Document_CORSIA%20Eligible%20Fuels_LCA%20Methodology.pdf.

[39] Pehnelt, G., and C. Vietze. 2013. "Recalculating GHG Emissions Saving of Palm Oil Biodiesel." Environment, Development and Sustainability 15: 429–79. https://doi.org/10.1007/s10668-012-9387-z.

[40] "Anaerobic Treatment of Palm Oil Mill Effluents: Potential Contribution to Net Energy Yield and Reduction of Greenhouse Gas Emissions from Biodiesel Production." 2013. Journal of Cleaner Production. https://www.sciencedirect.com/science/article/abs/pii/S0959652613005179.

[41] "Production of Palm Oil with Methane Avoidance at Palm Oil Mill: A Case Study of Cradle-to-Gate Life Cycle Assessment." n.d. ResearchGate record. https://www.researchgate.net/publication/282592962_Production_of_Palm_Oil_with_methane_avoidance_at_Palm_Oil_Mill_A_case_study_of_cradle-to-gate_life_cycle_assessment.

[42] U.S. Patent 8,282,815. n.d. "Method of Converting Feedstocks from Renewable Sources to Good-Quality Diesel Fuel Bases Using a Zeolite Catalyst without Intermediate Gas-Liquid Separation." https://image-ppubs.uspto.gov/dirsearch-public/print/downloadPdf/8282815.

[43] U.S. Patent 8,828,217. n.d. "Gas and Liquid Phase Hydroprocessing for Biocomponent Feedstocks." https://image-ppubs.uspto.gov/dirsearch-public/print/downloadPdf/8828217.

[44] U.S. Patent 11,802,247. n.d. "Process and Plant for Producing Hydrocarbons with Reduced CO2-Footprint and Improved Hydrogen Integration." https://image-ppubs.uspto.gov/dirsearch-public/print/downloadPdf/11802247.

[45] U.S. Patent 8,816,141. n.d. "Reducing Hydrogen Consumption in Hydrotreating of Biocomponent Feeds." https://image-ppubs.uspto.gov/dirsearch-public/print/downloadPdf/8816141.

[46] International Energy Agency. 2024. "GHG Emissions of Hydrogen and Its Derivatives." In Global Hydrogen Review 2024. Paris: IEA. https://www.iea.org/reports/global-hydrogen-review-2024/ghg-emissions-of-hydrogen-and-its-derivatives.

[47] International Energy Agency. 2023. Towards Hydrogen Definitions Based on Their Emissions Intensity. Paris: IEA. https://www.iea.org/reports/towards-hydrogen-definitions-based-on-their-emissions-intensity/executive-summary.

[48] Bauer, Christian, Karin Treyer, Cristina Antonini, et al. 2022. "On the Climate Impacts of Blue Hydrogen Production." Sustainable Energy & Fuels 6 (1): 66–75. https://pubs.rsc.org/en/content/articlepdf/2022/se/d1se01508g.

[49] Howarth, Robert W., and Mark Z. Jacobson. 2021. "How Green Is Blue Hydrogen?" Energy Science & Engineering 9 (10): 1676–87. https://www.ourenergypolicy.org/resources/how-green-is-blue-hydrogen/.

[50] Howarth, Robert W., and Mark Z. Jacobson. 2022. "Reply to Comment on 'How Green Is Blue Hydrogen?'" Energy Science & Engineering. https://doi.org/10.1002/ese3.1154.

[51] Edwards, R., M. Padella, J. Giuntoli, R. Koeble, A. O'Connell, C. Bulgheroni, and L. Marelli. 2017. Definition of Input Data to Assess GHG Default Emissions from Biofuels in EU Legislation. Version 1c. JRC Science for Policy Report EUR 28349 EN. Luxembourg: Publications Office of the European Union. https://publications.jrc.ec.europa.eu/repository/bitstream/JRC104483/kj1a28349enn.pdf.

[52] Edwards, R., M. Padella, J. Giuntoli, R. Koeble, A. O'Connell, C. Bulgheroni, and L. Marelli. 2019. Definition of Input Data to Assess GHG Default Emissions from Biofuels in EU Legislation. Version 1d. JRC Science for Policy Report. Luxembourg: Publications Office of the European Union. https://publications.jrc.ec.europa.eu/repository/bitstream/JRC115952/kj3a28349enn_online.pdf.

[53] Kourkoumpas, Dimitrios-Sotirios, Angeliki Sagani, Anastasios Vallianatos, Spyros Kiartzis, Sotirios Karellas, Vassilis Dedoussis, and Panagiotis Grammelis. 2024. "Life Cycle GHG Emission Reduction of Hydrotreated Vegetable Oil Integration in an Industrial Petroleum Refinery." Green Technologies and Sustainability 2 (2): 100076. https://doi.org/10.1016/j.grets.2024.100076.

[54] "Use of Waste Vegetable Oil for Hydrotreated Vegetable Oil Production with High-Temperature Electrolysis as Hydrogen Source." 2020. Fuel. https://www.sciencedirect.com/science/article/abs/pii/S001623612030987X.

[55] Hamelinck, Carlo. 2024. "GHG Calculations for Renewable Transport Fuels in the EU." Presentation, studio Gear Up, SGAB webinar, August 28. https://www.studiogearup.com/wp-content/uploads/2024/09/24_0828_sGU_Hamelinck_Carbon-Accounting-SGAB-Webinar.pdf.

[56] Fargione, Joseph, Jason Hill, David Tilman, Stephen Polasky, and Peter Hawthorne. 2008. "Land Clearing and the Biofuel Carbon Debt." Science 319 (5867): 1235–38. https://energyjustice.net/ethanol/ghg/2008-Fargione-Science-1235-8.pdf.

[57] Xu, H., U. Lee, and M. Wang. 2020. "Life-Cycle Energy Use and Greenhouse Gas Emissions of Palm Fatty Acid Distillate Derived Renewable Diesel." Renewable and Sustainable Energy Reviews 134: 110144. https://doi.org/10.1016/j.rser.2020.110144.

[58] Martin, Jeremy. n.d. "Lessons Learned from 18 Years of Arguing about Indirect Land Use Change." Union of Concerned Scientists blog. https://blog.ucs.org/jeremy-martin/lessons-learned-from-18-years-of-arguing-about-indirect-land-use-change/.

[59] California Air Resources Board. n.d. "Attachment B: Indirect Land Use Change." LCFS fuel pathway B0520. https://ww2.arb.ca.gov/sites/default/files/classic/fuels/lcfs/fuelpathways/comments/tier2/b0520_attachment_b.pdf.

[60] International Council on Clean Transportation. 2017. "How Rapeseed and Soy Biodiesel Drive Oil Palm Expansion." ICCT Briefing, July. https://theicct.org/sites/default/files/publications/Oil-palm-expansion_ICCT-Briefing_27072017_vF.pdf.

[61] "Understanding Uncertainty in Market-Mediated Responses to US Oilseed Biodiesel Demand: Sensitivity of ILUC Emission Estimates to GLOBIOM Parametric Uncertainty." 2025. PubMed Central, PMC11741104. https://pmc.ncbi.nlm.nih.gov/articles/PMC11741104/.

[62] Clean Fuels Alliance America and CABA. n.d. "Clean Fuels and CABA Comments to CARB on 15-Day Package." https://cleanfuels.org/wp-content/uploads/Clean-Fuels-and-CABA-Comments-to-CARB-on-15-Day-Package-Final.pdf.

[63] California Air Resources Board. n.d. CARB LCFS Fuel Pathway Report: Renewable Diesel (B0477). https://ww2.arb.ca.gov/sites/default/files/classic/fuels/lcfs/fuelpathways/comments/tier2/b0477_report.pdf.

[64] "The Carbon Footprint of Biofuels: Can We Shrink It Down to Size in Time?" 2008. Environmental Health Perspectives. PubMed Central, PMC2430251. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2430251/.

[65] "UCO Provides Greater Carbon Reduction Potential than Plant Oils across Diverse Biodiesel Production Technologies." 2026. Journal of Cleaner Production. https://www.sciencedirect.com/science/article/abs/pii/S0959652626003227.

[66] California Air Resources Board. n.d. CARB LCFS Fuel Pathway Report: Renewable Diesel (B0940). https://ww2.arb.ca.gov/sites/default/files/classic/fuels/lcfs/fuelpathways/comments/tier2/b0940_report.pdf.

[67] Washington State Department of Ecology. 2022. Indirect Land Use Conversion for Washington Clean Fuels Standard. https://ecology.wa.gov/getattachment/be3e311f-34de-4001-a055-b6dd07d25ead/iLUC20220404.pdf.

[68] International Civil Aviation Organization. n.d. CORSIA Supporting Document: CORSIA Eligible Fuels, Life Cycle Assessment Methodology. Version 3. https://www.icao.int/sites/default/files/sp-files/environmental-protection/CORSIA/Documents/CORSIA_Supporting_Document_CORSIA%20Eligible%20Fuels_LCA_Methodology_V3.pdf.

[69] Transport & Environment. n.d. "UCO (Unknown Cooking Oil): High Hopes on Limited and Suspicious Materials." https://www.transportenvironment.org/articles/uco-unknown-cooking-oil-high-hopes-on-limited-and-suspicious-materials.

[70] Irish Examiner. n.d. "EU Urged to Act on 'Obvious' Biofuel Fraud." https://www.irishexaminer.com/farming/arid-41558911.html.

[71] Malins, Chris. 2023. The Fat of the Land. Cerulogy, for Transport & Environment. May. https://www.cerulogy.com/wp-content/uploads/2023/07/Cerulogy_Fat-of-the-land_May2023.pdf.

[72] Chemical & Engineering News. 2012. "Oleochemical Industry Fears Loss of Key Raw Material to Biodiesel." C&EN 90 (47). https://cen.acs.org/articles/90/i47/Oleochemical-Industry-Fears-Loss-Key.html.

[73] Malins, Chris. 2017. Waste Not Want Not: Understanding the Greenhouse Gas Implications of Diverting Waste and Residual Materials to Biofuel Production. Cerulogy, for the International Council on Clean Transportation. August. https://theicct.org/wp-content/uploads/2021/06/Waste-not-want-not_Cerulogy-Consultant-Report_August2017_vF.pdf.

[74] "Renewable Diesel from Oils and Animal Fat Waste: Implications of Feedstock, Technology, Co-products and ILUC on Life Cycle GWP." 2020. ResearchGate record. https://www.researchgate.net/publication/342018298_Renewable_diesel_from_oils_and_animal_fat_waste_implications_of_feedstock_technology_co-products_and_ILUC_on_life_cycle_GWP.

[75] American Soybean Association. n.d. "Advocating for Soy Biomass-Based Diesel in California." https://soygrowers.com/carb-lcfs/.

[76] International Council on Clean Transportation. n.d. "Kein Cap? There's More than Meets the Eye with the EU's Waste Fats and Oils Limit." https://theicct.org/kein-cap-theres-more-than-meets-the-eye-with-the-eus-waste-fats-and-oils-limit/.

[77] Transport & Environment. 2024. The Advanced and Waste Biofuels Paradox. July. https://uploads.transportenvironment.org/production/files/202407_TE_advanced_biofuels_report-1.pdf.

[78] Flach, B., S. Lieberz, and S. Bolla. 2019. EU Biofuels Annual 2019. USDA Foreign Agricultural Service GAIN Report. The Hague. https://apps.fas.usda.gov/newgainapi/api/report/downloadreportbyfilename?filename=Biofuels+Annual_The+Hague_EU-28_7-15-2019.pdf.

[79] Environment, Land & Resources. 2025. "California Low Carbon Fuel Standard Amendments Take Effect." August. https://www.globalelr.com/2025/08/california-low-carbon-fuel-standard-amendments-take-effect/.

[80] AgWeb. n.d. "CARB Revises Proposal to LCFS Regulations." Pro Farmer analysis. https://www.agweb.com/markets/pro-farmer-analysis/carb-revises-proposal-lcfs-regulations.

[81] Biobased Diesel Daily. n.d. "California Updates LCFS Program, Moves Forward with 20% Cap on Biofuels from Crop Oils." https://www.biobased-diesel.com/post/california-updates-lcfs-program-moves-forward-with-20-cap-on-biofuels-from-crop-oils.

[82] Christianson PLLP. n.d. "California Finalizes Major Changes to Low Carbon Fuel Standard (LCFS)." https://www.christiansoncpa.com/california-finalizes-major-changes-to-low-carbon-fuel-standard-lcfs/.

[83] Congressional Research Service. n.d. The Section 45Z Clean Fuel Production Credit. IF12502. https://www.congress.gov/crs-product/IF12502.

[84] American Farm Bureau Federation. n.d. "45Z Clean Fuel Production Credit." Market Intel. https://www.fb.org/market-intel/45z-clean-fuel-production-credit.

[85] Crux. n.d. "What the Updated 45ZCF-GREET Model Means for Clean Fuel Producers." https://www.crux.com/blog/what-the-updated-45zcf-greet-model-means-for-clean-fuel-producers.

[86] Reuters. 2024. "France, Germany Urge Tougher EU Checks on Biofuel Imports in Fraud Probe." 31 May. https://www.swissinfo.ch/eng/france,-germany-urge-tougher-eu-checks-on-biofuel-imports-in-fraud-probe/79248810.

[87] S&P Global Commodity Insights. 2025. "Indonesia Restricts UCO, Palm Oil Residue Exports Ahead of B40 Biodiesel Mandate." 8 January. https://www.spglobal.com/energy/en/news-research/latest-news/agriculture/010825-indonesia-restricts-uco-palm-oil-residue-exports-ahead-of-b40-biodiesel-mandate.

[88] U.S. Department of Agriculture, Foreign Agricultural Service. 2025. Indonesia Curbing Palm Waste Exports: Discouraging CPO Mixture Practices. GAIN Report ID2025-0009. Jakarta.

[89] European Commission. 2025. "Commission Implementing Regulation (EU) 2025/261 of 10 February 2025 Imposing a Definitive Anti-Dumping Duty on Imports of Biodiesel Originating in the People's Republic of China." Official Journal of the European Union L. https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=OJ%3AL_202500261.

[90] Malins, Chris, and Cato Sandford. 2022. Animal, Vegetable or Mineral (Oil)? Exploring the Potential Impacts of New Renewable Diesel Capacity on Oil and Fat Markets in the United States. Cerulogy, January. https://www.cerulogy.com/wp-content/uploads/2022/01/Cerulogy_Animal-vegetable-mineral_Jan2022-1.pdf.