Quick answer: Anaerobic digestion is settled science and most plant work is engineering with a determinable outcome. In co-digestion, a core R&D activity may exist where a feedstock and sludge pairing has no established basis for predicting whether it can be loaded stably to a required gas and product specification — a biochemical methane potential assay measures ultimate methane yield in a bottle and does not answer that question. In biosolids, using a listed approved process and sampling batches against it is generally unlikely to be a core R&D activity on those facts, while determining whether an unproven route can reliably reach a grade's microbiological criteria on a particular sludge is a different activity. In landfill gas, wellfield balancing and prescribed emissions estimation are established practice, while an intervention such as a methanotrophic biocover with no site-specific basis may be. Each activity is assessed against the statutory tests on its own facts and the company self-assesses.
24 August 2026 — this article describes the current rules. The 2026-27 Federal Budget announced R&DTI changes; the ATO states the measure is not yet law, and industry.gov.au states the changes will apply to income years starting on or after 1 July 2028.
Anaerobic digestion has been making usable gas from sewage sludge since the nineteenth century. The biochemistry is textbook, the reactors are catalogue items, and most of what a utility or waste operator does with them is competent process engineering with a determinable outcome. When a digester takes a feedstock it has never seen, the genuinely unknown part is small, specific, and not where people usually point. This article locates it — in co-digestion, in biosolids treated to a reuse grade, and in landfill gas — under a registered research field of environmental engineering (ANZSRC 4011).
The Statutory Test, Stated Once
Eligibility is assessed activity by activity, not project by project. Under s 355-25(1) of the Income Tax Assessment Act 1997, core R&D activities are experimental activities whose outcome cannot be known or determined in advance on the basis of current knowledge, information or experience, but can only be determined by applying a systematic progression of work that is based on principles of established science and that proceeds from hypothesis to experiment, observation and evaluation, and leads to logical conclusions; and that are conducted for the purpose of generating new knowledge, including new knowledge in the form of new or improved materials, products, devices, processes or services (business.gov.au; ITAA 1997).
A supporting R&D activity must be directly related to core R&D activities (s 355-30(1)); and where an activity is of a kind listed in s 355-25(2), produces goods or services, or is directly related to producing goods or services, s 355-30(2) adds a dominant purpose bar (business.gov.au). That third limb is the one that reaches receival infrastructure, blending plant and commissioning on a works simultaneously making electricity and a saleable soil product. Everything else in this article is technical.
Microbial ecology, thermodynamics and reactor theory give any digestion investigation a settled body of principle to build a progression on, so the established-science limb is readily capable of being satisfied — though whether it is satisfied turns on how the particular progression is designed and recorded. The limb that bites is the first: digestion is so well characterised that most questions asked of it can be answered in advance, by a mass balance, a design standard, a vendor curve or a laboratory assay. The practitioner's work, then, is narrow and mostly done before the first tanker arrives — isolate the precise technical uncertainty, establish what a competent professional in the field could already have determined without experiment, define the experimental acts that would resolve what is left, and keep the contemporaneous record that distinguishes those acts from commissioning and production. See what does not qualify.
Why a BMP Number Does Not Predict What a Digester Will Do
The biochemical methane potential (BMP) assay is the standard first question asked of a candidate co-substrate: incubate it with a dilute, well-adapted inoculum and excess buffer, and measure cumulative methane. It measures one thing well — the ultimate methane yield under non-limiting conditions. Five things it structurally cannot tell you:
Hydrogen partial pressure, and therefore propionate: Syntrophic oxidation of propionate to acetate, CO2 and H2 is endergonic under standard conditions — of the order of +76 kJ/mol — and proceeds only while hydrogenotrophic methanogens hold hydrogen partial pressure low; butyrate oxidation tolerates roughly an order of magnitude more. Propionate is the tightest link in the chain, and a digester in trouble announces it as propionate accumulating while acetate does not. A bottle at a substrate-to-inoculum ratio of 0.5 never approaches the acidogenic burst a full-scale slug feed produces, so it cannot expose that link.
Free ammonia, which is a pH and temperature quantity, not a nitrogen quantity: The inhibitor is un-ionised NH3, which crosses the cell membrane and protonates in the near-neutral cytoplasm, dissipating the proton gradient. Its fraction of total ammonia nitrogen is 1/(1 + 10(pKa − pH)), with pKa near 8.95 at 35 °C — so a 0.3-unit rise in digester pH roughly doubles free ammonia at unchanged total nitrogen. Acetoclastic methanogens are more sensitive than hydrogenotrophic ones, so a protein-rich co-substrate can push the community towards syntrophic acetate oxidation and the digester keeps running at apparent steady state on a much narrower margin. Buffered, dilute bottles do not reproduce that.
Foam, which needs three things at once: A surface-active species (long-chain fatty acids from fats, oils and grease; soluble protein; biosurfactants), hydrophobic filamentous organisms carried in from the activated-sludge plant that stabilise the bubble film, and a gas flux to make bubbles. Foam blinds the gas take-off and pressure-relief and — the consequence usually missed — cuts effective liquid volume, so real retention time falls below the design value still shown on the plan.
Gas quality, which moves in two directions at once: The methane fraction is set by the mean oxidation state of the carbon fed — lipids near 70% CH4, carbohydrate near 50% — so a fat-rich co-substrate raises methane content and shifts the Wobbe index the engine was mapped for. The same feedstock's sulfur-bearing amino acids mineralise to sulfide, and the H2S loading can rise by a multiple that consumes scrubber media and engine oil alkalinity far faster than budgeted. A BMP reports methane; it is not a gas-quality measurement.
What the digestate does next: Co-digestion can reduce dewaterability, raising polymer dose and lowering cake dryness, putting tonnes back on the road — sometimes a material share of the value of the extra gas, and no bottle assay indicates it.
None of which makes co-digestion research by default: where the operator has run the feedstock before, or the supplier can produce full-scale data for a comparable sludge and loading, the outcome is determinable in advance.
Biosolids Grades: Approved Processes and Demonstrated Alternatives
Biosolids reuse is governed by national guidance (NWQMS Guidelines for Sewerage Systems — Biosolids Management) and by state instruments. In South Australia biosolids carry both a Contamination Grade and a Stabilisation Grade, and the EPA's guidelines list approved processes for each: anaerobic digestion for a period of between 15 days at 35–55 °C and 60 days at 20 °C reaches Stabilisation Grade B, while Grade A requires composting, heat drying, pasteurisation, lime stabilisation with heating, or three years' storage. Vector-attraction reduction can be met by a process that reduces volatile solids by 38%, and the Grade A microbiological criteria are numeric: fewer than 100 E. coli per gram of total solids, and fewer than one Salmonella, one virus and one viable helminth ovum per 50 grams.
The sentence that matters for research sits under those tables: other processes will be approved provided it is demonstrated that they reliably achieve the grade's microbiological requirements. Validation sampling on a listed process to show a batch complies reads naturally as an activity associated with complying with statutory requirements or standards — a kind s 355-25(2)(f) excludes from being a core activity, and an exclusion that does not turn on the result being predictable, since a batch can fail. Determining whether a route with no established basis can reliably achieve the criteria on a particular sludge is a different activity, assessed on its own facts; the surrounding plant and sampling then have to be worked through against s 355-30 as set out above.
Landfill Gas: The Same Biology, in a Reactor Nobody Can Mix
A landfill is a digester with no mixing, no temperature control, an unknown charge and a decades-long time constant. At larger Australian sites — including those accepting about 75% of municipal solid waste — gas is collected and combusted for its energy value, usually generating electricity for sale (DCCEEW), and generation is estimated with first-order decay models whose rate constants and methane potentials are set by waste stream and climate zone under the reporting methodology (DCCEEW). Applying those models is estimation, not experiment, and the physical work — wellfield balancing between air ingress on one side and fugitive loss through the cover on the other — is established practice with a determinable outcome.
The narrower case is an intervention with no established site-specific basis. Take a closed cell under a clay cover, with the wellfield already balanced to the limit that wellhead methane and oxygen readings allow, still losing a residual flux through the surface. Whether a compost-amended methanotrophic biocover can oxidise a defined share of that residual flux — on this waste age and composition, this climate zone, this cover geometry and this moisture regime — is not answered by the decay model, which estimates generation rather than what a surface layer does to it, nor by published biocover results obtained on other covers. A testable hypothesis states the layer specification and the share of residual flux it is expected to oxidise, sustained across a full seasonal cycle at unchanged wellfield vacuum. Held: extraction rate, vacuum setpoint, cell geometry. Varied: amendment ratio, layer depth, and the moisture and compaction regime. Measured: surface flux by static chambers on a fixed grid, methane and oxygen concentration profiles through the layer depth, temperature, and the carbon-isotope shift between the gas entering the layer and the gas leaving it, which is what separates oxidation from dilution by air ingress. Outside that programme sit the routine surface scans, the wellhead balancing the site does anyway, and the NGER estimate itself.
A Worked Example (Hypothetical and Illustrative Only)
Invented to show where the boundary falls. It is not a real project and says nothing about whether any actual claim would be accepted.
A regional South Australian utility runs a 45,000 EP works with two 2,400 m³ mesophilic digesters, one in service at 36 °C, fed blended primary and waste activated sludge at 4.8% dry solids and 76% volatile — about 2,260 kg VS/d — at 62 m³/d, so about 39 days' retention. A nearby poultry processor offers its dissolved-air-flotation float, a high-fat, high-protein material, under a trade-waste arrangement.
Baseline: Over 12 weeks before changes: biogas 820 Nm³/d at 63.4% CH4 (520 Nm³ CH4/d, ~230 L CH4/kg VS fed); volatile solids destruction 47%; total ammonia nitrogen 980 mg/L at pH 7.35 (~24 mg/L free ammonia); propionate 85 mg/L; H2S 620 ppmv (vs 900 ppmv scrubber design); cake 18.5% DS at 9.5 kg polymer/tDS. BMP screening returned 610 ± 34 L CH4/kg VS for float vs 285 for sludge. At 15% co-substrate VS (~400 kg VS/d), assay projected 763 Nm³ CH4/d (+47%).
Target, recorded before trial: At 15% co-substrate VS, sustain at least a 35% increase in daily methane over baseline mean (≥ 702 Nm³ CH4/d) while holding total VFA < 1,500 mg/L as acetic acid, propionate < 400 mg/L, intermediate-to-partial alkalinity ratio < 0.35, foam < 300 mm above operating level, VS destruction ≥ 38%, and retention time ≥ 15 days at 35–55 °C (Stabilisation Grade B).
Held & varied: Held: 36.0 ± 0.4 °C, primary:WAS at 60:40 on VS, gas mixing at 210 Nm³/h, 8 feeds/day on fixed clock, 1 NATA laboratory, 07:00 sampling, 1 polymer/centrifuge setting. Varied: co-substrate share of VS load (0, 5, 10, 15, 20%), step interval, introduction mode and trace-element dose.
Phase 1 — Single daily slug (Failed): Stepped straight to 10%. Gas rose to 990 Nm³/d for 9 days then fell. By day 16 propionate rose from 85 to 1,180 mg/L, alkalinity ratio to 0.41, gas fell to 730 Nm³/d. Free ammonia was 34 mg/L (hypothesis not supported). Cutting co-substrate on day 19 showed propionate took 58 days to fall below 400 mg/L (longer than 44-day hydraulic washout), proving continued production. Ruled out: single daily slug feed.
Phases 2 to 4: Pre-blending float across 8 daily feeds (Phase 2) held 10% stably. At 15%: 1,085 Nm³/d biogas (726 Nm³ CH4/d, +40%), but propionate sat at 380–430 mg/L (over ceiling). Micronutrient dosing (Phase 3) did not reduce propionate. Injecting float upstream of heat exchanger + 20 min gas-mixing pause after feed (Phase 4) held foam at 160 mm and brought propionate to 340 ± 25 mg/L over 6 weeks, meeting every criterion.
Results & boundaries: Methane fraction rose to 66.9% (engine map re-tuned); H2S rose to 1,760 ppmv (scrubber media life cut from 14 to 5 weeks); cake fell to 16.2% DS (polymer rose to 13.1 kg/tDS); VS destruction rose to 52%; retention held at ~35 days. Candidate experimental activity runs from hypothesis/pass-criteria formulation through step-loading trials to characterisation of the Phase 4 envelope.
Outside candidate boundary: Receival bay and blend pumps, trade-waste agreement, routine licence sampling, steady operation at 15%, and replicating at a second plant (each needing supporting-activity analysis under s 355-30, including dominant purpose).
What Is Generally Unlikely to Be Core on Those Facts
Activity
Why
Sizing a digester, mixing system or gas holder from established design loadings
Established method, outcome determinable in advance
Running a BMP assay to characterise a candidate feedstock
May be routine testing and analysis under s 355-25(2)(f)(iii) where undertaken for that purpose; otherwise assessed on its own facts
Validation sampling to show a batch meets a Stabilisation Grade by a listed approved process
Activity associated with complying with statutory requirements or standards
Wellfield balancing, and applying first-order decay models to estimate landfill gas
Established practice and prescribed estimation, not experiment
Repeating a proven co-digestion recipe at a second plant with comparable sludge
Outcome determinable from the first plant's data
Where an RSP Fits
An RSP is a scientific or technical service provider registered in specific fields that a company can engage to conduct R&D activities on its behalf (business.gov.au). On a digestion programme that work sits upstream of the first tanker: framing what is actually unknown, designing the step-loading and hold sequence so a failure is interpretable, and specifying the analytical suite and sampling frequency at the resolution the stability measures require (record keeping). There is also a point that matters for smaller operators: RSP-conducted eligible R&D activities can be claimed even where the usual $20,000 R&D expenditure threshold is not met, and using an RSP does not guarantee eligibility — you still self-assess. In mechanism, where a company's total notional R&D deductions for an income year come to less than $20,000, s 355-100(2) substitutes a different base for working out the offset, and the first kind of expenditure that base is made of is expenditure incurred to a registered research service provider that is not an associate of the R&D entity, for services within a research field for which the provider is registered. The offset tiers are covered in our article on the refundable and non-refundable offset; see also claiming R&D under $20,000 and the R&D Tax Incentive in Adelaide.
Talk to Ignition Research before the first co-substrate tanker arrives, while the trial can still be separated from normal operation. As a Registered Research Service Provider at Lot Fourteen in Adelaide working in environmental engineering, we design and conduct the experimental programme and produce the technical record while the work is happening. We are not a registered tax agent: your company self-assesses and remains responsible for its own claim. Get in touch.
Frequently Asked Questions
Q: Is anaerobic co-digestion an eligible R&D activity?
A: Not by default. Adding a well-characterised co-substrate at a documented loading, where comparable full-scale operating data exists, applies an established method with a determinable outcome. A core R&D activity may exist in the narrower case where a feedstock and sludge pairing has no established basis for predicting whether it can be loaded stably to a required gas and product specification, and the answer comes only from a systematic progression of instrumented trials, based on principles of established science and conducted to generate new knowledge. You self-assess.
Q: Why can't a BMP test not predict full-scale biogas yield?
A: A BMP measures ultimate methane potential in a dilute, buffered batch bottle at a low substrate-to-inoculum ratio. A continuous digester adds constraints the bottle removes: hydrolysis kinetics against a finite retention time, hydrogen partial pressure and the propionate bottleneck under transient loading, free ammonia and long-chain fatty acid inhibition at real concentrations, foaming — for which a serum bottle reproduces neither the gas flux, the mixing energy, the rheology, the surface geometry nor the continuous solids loading of a full-scale digester — and downstream effects on gas quality and dewaterability. The BMP is a valid measurement of one variable, not a prediction of plant behaviour.
Q: Is digester foaming a research problem or an operating problem?
A: Usually an operating problem. Where the foam-forming mechanism and the effective counter-measures are known for the sludge in question, tuning feed pattern, mixing or antifoam dose is established practice. It may become experimental where a new feedstock produces foam that established measures demonstrably do not control, and determining whether any operating configuration can hold it within a defined limit requires a systematic progression of instrumented trials. That is assessed on the activity's own facts.
Q: Is treating biosolids to Stabilisation Grade A an R&D activity?
A: Using a listed approved process — composting, thermal drying, pasteurisation, lime stabilisation with heating or long-term storage — and sampling to demonstrate a batch complies is generally unlikely to be a core R&D activity on those facts: activities associated with complying with statutory requirements or standards are excluded from being core by s 355-25(2)(f). Guidance contemplates alternative processes being approved where it is demonstrated they reliably achieve the microbiological criteria, and work directed at determining whether an unproven route can do so on a particular sludge is a different activity assessed on its own facts.
Sources & Further Reading
legislation.gov.au — Income Tax Assessment Act 1997 — Div 355, incl. ss 355-25 and 355-30
waterquality.gov.au — Guidelines for Sewerage Systems: Biosolids Management
EPA South Australia — Guidelines for the safe handling and reuse of biosolids in South Australia
DCCEEW — Energy recovery, national waste and resource recovery reporting
Related: environmental engineering research field · what does not qualify · what an RSP is · refundable vs non-refundable offset · R&D Tax Incentive in Adelaide · more Insights
This article is general information from a Registered Research Service Provider about the R&D Tax Incentive. It is not tax, legal or financial advice; eligibility depends on your circumstances and you should self-assess and seek your own advice.
Thinking about a project like this?
If you're weighing up an AI, software or technical improvement project and can't tell yet whether it's implementation or research, start with a quick read on where it sits.