Quick answer: A remediation trial is not automatically R&D. Under s 355-25(1) of the ITAA 1997 a core R&D activity needs an outcome that could not be known or determined in advance and could only be determined by a systematic progression of work, conducted to generate new knowledge. In situ, that usually turns on whether reagent delivery, persistence and post-treatment rebound in this particular subsurface can be predicted from bench results; sizing a known technique generally cannot clear that bar. Clearing it is not the end of the analysis: s 355-25(2)(f) separately excludes activities associated with complying with statutory requirements or standards, including routine testing and analysis of soils — and most remediation happens because a regulator requires it. Each activity is assessed separately against the statutory tests, and the company self-assesses its own claim.
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.
Two pilot trials can look identical from the site fence. Both put a rig over a former industrial block, drill injection points, push reagent into an aquifer and sample monitoring wells for six months. One is answering a question nobody could answer beforehand. The other is measuring a number a competent professional could have estimated from published design data, to size a job.
This article is about in-situ remediation trial design on general contaminated land — former industrial and commercial sites, service-station and dry-cleaning legacies, urban infill redevelopment. Mine-site tailings and mine water management are a different technical and regulatory setting, covered separately in our mining and METS Insight.
Regulatory and environmental assessment boundary
One boundary first. This article is only about the R&D Tax Incentive question — whether a technical outcome was genuinely not knowable in advance. It does not endorse any technology's effectiveness or safety. Assessment against regulatory standards and human-health risk assessment are matters for accredited environmental and health professionals working within the applicable framework — the Assessment of Site Contamination NEPM and the relevant state regulator's requirements — not for this article.
What Stops the Beaker Predicting the Aquifer
Scale is the wrong dividing line: a bench study can be entirely predictable, and a full-scale injection can be the only way to resolve a question because the thing in doubt exists only at field scale. A trial run to size a known technique is measuring a parameter. A trial whose outcome genuinely cannot be determined in advance is testing a proposition. How many injection points, at what spacing, to deliver a design loading is engineering design when the technique, the contaminant and the matrix all sit inside a documented application envelope. Four recurring mechanisms sit behind most of the gap between the two cases:
Reagent demand is a rate, not a number: For in-situ chemical oxidation the natural oxidant demand of the soil — oxidant consumed by native organic matter and reduced iron and manganese minerals rather than by the contaminant — commonly dominates the loading, and it is kinetically staged. Short- and long-duration batch measurements on the same composite can differ substantially, because the fast and slow fractions of the matrix behave as different populations. Design from the short figure and the field can be under-dosed; design from the long one and the job can be uneconomic. Which figure governs at field residence times, in grams of oxidant per kilogram of soil, is not always derivable from either — and the residence time itself is set by how long reagent stays in contact with the mass, which is a delivery question rather than a chemistry one.
Radius of influence is a delivery outcome, not a property of the reagent: How far injected fluid actually travels depends on hydraulic conductivity contrast, injection pressure and rate, fluid viscosity and density, and whether the formation accepts flow evenly or opens a preferential path. A conservative tracer injected with the reagent and tracked in surrounding wells measures the distribution rather than assuming it — and buried services trenches, old foundations and backfill routinely produce a distribution that is not radial at all. A design that quotes a single radius has already assumed the answer to the question the trial exists to ask.
Rebound reports last, and it can reverse the reading: Concentrations fall during and after treatment, then commonly rise again as mass diffuses back out of low-permeability silt and clay lenses into the transmissive sand, or desorbs from soil organic carbon. A 90-day result showing a large reduction and a 270-day result showing most of it returned are the same trial. Which of the two readings a design is judged on, and against what ratio, therefore has to be settled before the first injection rather than after the last round of sampling — otherwise the evaluation limb has nothing fixed to evaluate against.
Heterogeneity converts a reaction-limited problem into a mass-transfer-limited one: A batch reactor holds sieved, homogenised soil in complete contact with reagent, so it measures reaction kinetics. In the ground, reagent travels through the transmissive fraction while much of the mass sits in the low-permeability fraction it barely enters, and the rate that governs is diffusion across that interface rather than the destruction rate of the compound. The bench answers whether the chemistry destroys the contaminant; the field asks whether the chemistry ever meets it. Where the second question governs, a bench result of near-complete destruction carries little predictive weight, and that is where a genuine technical unknown tends to live.
What the Legislation Asks, and the Exclusions That Press Hardest Here
Section 355-25(1) of the Income Tax Assessment Act 1997 defines core R&D activities as 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 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).
Redox chemistry, groundwater hydraulics, advection-dispersion and diffusion transport and microbial reductive dechlorination supply a settled body of principle a progression of work here can be based on. It is the hypothesis-to-evaluation limb that catches remediation programmes: an inject-sample-inject loop that never fixes a hypothesis in advance and never evaluates against it can be careful and well documented, and still not be a systematic progression of work in the sense the section uses.
Closer to an outcome that could not be determined in advance
Ordinarily design, delivery or characterisation
Whether any delivery configuration puts reagent in contact with mass held in a specific low-permeability structure, where tracer data show distribution is not radial
Calculating point spacing from a radius of influence already established for that formation and method
Whether the rebound asymptote is bounded at all, where back-diffusion behaviour is not described by available data for that contaminant and lithology
Monitoring rebound after a treatment whose performance in that setting is documented
Whether a reagent survives long enough at field residence times in a particular geochemistry — buffering, competing demand, mineral fouling
Measuring natural oxidant demand to a standard method to set a loading, inside a published envelope
The right-hand column is not a judgement about difficulty. Characterising a plume to the standard a regulator expects is demanding, and is generally unlikely to be a core R&D activity on those facts, subject to the activity's own facts and the statutory tests.
Paragraph (f) — Statutory requirements, standards and soils testing
Section 355-25(2)(f) identifies certain activities associated with complying with statutory requirements or standards as excluded from being core R&D activities, including one or more of the following: (i) maintaining national standards; (ii) calibrating secondary standards; and (iii) routine testing and analysis of materials, components, products, processes, soils, atmospheres and other things. Paragraph (f)(iii) names soils expressly; a remediation programme generates a great deal of exactly that, and the project is usually carried out because a regulator requires it, which engages the opening words of (f) as well. Whether an activity is "associated with" complying with a requirement or standard is a question of fact assessed activity by activity. What can be said is narrower: the same borehole on the same day can yield a compliance sample and a measurement taken against a hypothesis fixed in advance, and those are different activities.
Section 355-30(1) makes supporting R&D activities those directly related to core R&D activities. Section 355-30(2) adds a further bar for listed kinds: if an activity (a) is an activity referred to in s 355-25(2); or (b) produces goods or services; or (c) is directly related to producing goods or services; it is a supporting R&D activity only if it is undertaken for the dominant purpose of supporting core R&D activities (business.gov.au). That matters here because a contractor selling remediation is producing a service, while a landowner cleaning up its own site may not be, and the drilling, reagent supply, injection, spoil handling and reporting that surround a candidate experiment are typically directly related to producing that service. Each activity is tested on its own, never the project or the site as a whole.
A Worked Example: A Chlorinated Solvent Source Zone Under an Infill Site
Hypothetical and illustrative. The numbers show what a worked investigation looks like. They are not a statement that the activity would be eligible, not a performance claim for any technology, and not clean-up criteria.
Baseline: Former metal-finishing works in inner Adelaide. Trichloroethene (TCE) in source-area groundwater: 42,000 µg/L, and 3,800 µg/L at 40 m downgradient. Treatment interval is 3–7 m below ground in a sand of ~5 m/day, interbedded with silty clay lenses 0.2–0.6 m thick (~10⁻³ m/day, ~30% of interval by thickness). Earlier programmes on homogeneous sand achieved 85–95% reduction; none had this lens structure.
Targets (trial performance criteria, not clean-up criteria): Evaluated across 5 monitoring points on radials per cell: ≥ 90% reduction in dissolved TCE by day 90 in cell mean; day-270 rebound ratio (day-270 conc / day-90 conc) ≤ 1.5; tracer detected at ≥ 2.5 m from injection point in ≥ 4 of 5 points; ≤ 20% loss of specific capacity at injection well.
Held constant & varied: Held: injection interval (4.0–6.5 m), well construction, bromide tracer (200 mg/L), sampling schedule (days 0, 7, 30, 90, 180, 270), laboratory, lithology block. Varied: reagent system (alkaline-activated persulfate, chelated-iron-activated persulfate, colloidal activated carbon + micro-scale ZVI amendment), oxidant loading (8, 16, 26 kg/m³), injection mode (gravity vs 150 kPa pressurised), and event structure (single slug vs two events 30 days apart ± day-60 amendment).
Sequence & trials:
Bench phase: Batch NOD measured 2.1 g/kg (48 h) and 6.4 g/kg (30 d). Persulfate destroyed 98% TCE at 16 g/kg (~26 kg/m³ at 1.6 t/m³ dry bulk density).
Field trial 1 — Pressurised single slug (26 kg/m³, Failed): Bromide reached ≤ 1.1 m in 4 of 5 points, and 4.5 m in the 5th (channelled along a former services trench). Persulfate was undetectable by day 11. TCE fell 61% by day 30 and rebounded to 74% of baseline by day 180. Showed: Delivery, not reactivity, was the binding constraint; pressurised injection did not distribute radially in this stratigraphy.
Field trial 2 — Gravity-fed two-event split (8 kg/m³ × 2, tighter spacing): Tracer reached 2.6 m in 4 of 5 points; specific capacity loss 6%. TCE fell 93% by day 90, but rebounded to 38% of baseline by day 270 (rebound ratio 5.4). Porewater from clay-lens cores held ~3× higher TCE than sand, confirming low-permeability back-diffusion.
Field trial 3 — Three concurrent cells (gravity-fed split): Cell A (8 kg/m³ × 2 + day-60 colloidal carbon amendment to intercept back-diffusion) reached 94% day-90 reduction, day-270 rebound ratio 1.2, capacity loss 9%. Cell B (control, no amendment) reached 92% reduction and 4.9 rebound ratio. Cell C (16 kg/m³ × 2 + amendment) reached 95% reduction, 1.3 rebound ratio, but 14% capacity loss.
The result actually reached: Cell A met all 4 criteria. Comparing Cell A to Cell B isolated the day-60 interception amendment as the true mechanism bounding rebound (1.2 vs 4.9). Doubling oxidant (Cell C) bought only 1% extra reduction while increasing well fouling. Rebound ratio correlated with clay lens thickness in the screened interval (1.05 at 0.35 m thickness up to 1.4 at 1.1 m thickness).
A candidate boundary: Candidate core activity spans the dated unknown and search through to evaluation of Field Trial 3. Preceding plume delineation, regulatory reporting, compliance/validation testing, full-scale deployment, and soil disposal sit outside, requiring separate supporting-activity analysis under s 355-30(2). Candidate core activity, subject to self-assessment.
Where an RSP Fits
Research Service Providers are organisations registered to provide scientific or technical services to companies conducting R&D, in specific fields of research (business.gov.au), including pollution and contamination. What an RSP contributes to a remediation programme is experimental structure: writing down the unknown and the acceptance criteria before the rig arrives, and designing a sequence with the comparator that separates a delivery failure from a chemistry failure rather than confounding the two.
One point on the money side is specific to RSPs: Eligible R&D expenditure incurred to a registered research service provider may be treated differently where the usual $20,000 R&D expenditure threshold is not met, subject to the requirements in s 355-100(2) and the underlying activities satisfying the R&D eligibility tests (business.gov.au) — see claiming R&D under $20,000 — and using an RSP does not guarantee eligibility — you still self-assess. In mechanism, where an R&D entity's total notional deductions for the year fall below $20,000, s 355-100(2) works the offset out on a substituted base, the first kind of expenditure in it being 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. Offset rates, tiers and how they are worked out are covered in refundable vs non-refundable.
Talk to Ignition Research before the pilot is mobilised — as a Registered Research Service Provider at Lot Fourteen in Adelaide, we help site owners and remediation contractors define what a trial is actually testing and design a sequence capable of resolving it. 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 a remediation pilot trial eligible for the R&D Tax Incentive?
A: Not as a category. It turns on whether the particular activity meets the core test set out above and how the exclusions apply. What separates the two pilots in practice is the record made at the time: a dated statement of the unknown; a prior-art search showing that the contractor's own data and the available published literature did not answer it for this contaminant, lithology and delivery method; a hypothesis with acceptance and failure criteria fixed before mobilisation; the variables held constant and varied, with the comparator that makes the variation readable; and a written evaluation against those criteria, including where the trial failed. A pilot run to size a technique already documented for that contaminant and matrix is generally unlikely to be a core R&D activity on those facts, subject to the activity's own facts and the statutory tests. You self-assess.
Q: Is a bench-scale treatability study an R&D activity?
A: Scale does not decide it. A batch test confirming published reaction chemistry at published loadings is characterisation. A bench sequence designed to resolve a question the literature leaves open — for example, how natural oxidant demand partitions between fast and slow fractions in a matrix outside the described range, where that partition governs the field design — can sit closer to the core test. The activity is assessed on its own facts, not on the equipment used.
Q: Is validation sampling routine testing under s 355-25(2)(f)?
A: Section 355-25(2)(f) excludes activities associated with complying with statutory requirements or standards, including (iii) routine testing and analysis of materials, components, products, processes, soils, atmospheres and other things — soils are named expressly. Validation and compliance sampling analysed against guideline values sits close to that subparagraph. "Associated with" is broad, and whether a particular sampling activity falls inside the exclusion is a question of fact assessed activity by activity; the company self-assesses.
Q: Does a remediation carried out under an EPA requirement automatically fail the standards exclusion?
A: It is not automatic, but the exclusion is engaged and has to be worked through, and the presence of a regulatory driver does not by itself establish that every activity within the programme is caught. The assessment is made against each activity rather than the project. Note too that s 355-30(2)(a) applies an additional restriction: where an activity is referred to in s 355-25(2), it can only qualify as a supporting R&D activity if the dominant-purpose test is satisfied. This is a self-assessment made on the facts, and the regulatory endpoints themselves are matters for accredited environmental professionals.
Sources & Further Reading
legislation.gov.au — Income Tax Assessment Act 1997 — Div 355, incl. ss 355-25 and 355-30
legislation.gov.au — National Environment Protection (Assessment of Site Contamination) Measure 1999
nepc.gov.au — Assessment of Site Contamination NEPM and its schedules
EPA South Australia — Site contamination: assessment and remediation
Related: pollution and contamination research field · mining, METS and mine-site water · what does not qualify · what an RSP is · claiming R&D under $20,000
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. It is not environmental, engineering or health advice, and it does not endorse the effectiveness or safety of any remediation technology.
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.