When Does Desalination Work — Membrane Fouling, Brine Concentrate and Energy Recovery — Become R&D Rather Than Plant Engineering?

24 August 2026 — this article describes the current rules. The 2026–27 Federal Budget announced R&DTI reforms that will apply to income years starting on or after 1 July 2028. Until then, the program continues to be administered under the current legislation, as stated by industry.gov.au .

A reverse-osmosis membrane is a passive object. What decides whether a plant holds its design flux is not the membrane itself but three things around it: the thin layer of water sitting against the membrane surface, where flux loss, salt passage and scale nucleation actually originate; the concentrate leaving the pressure vessel, which carries everything the membrane rejected and sets the volume and salinity the site has to dispose of; and the device that recovers pressure from that concentrate, which couples the two, because concentrate that mixes back into the feed raises the pressure needed for the same flux.

This article is about those three: what forms on the membrane surface on a particular feedwater, what happens to the concentrate, and how an energy recovery device behaves away from its rated duty. It is not about plant delivery, dosing control or SCADA — those sit elsewhere in our Insights. The question here is when one of those outcomes is genuinely not knowable in advance from what the field already knows.

The Statutory Test the Activity Has to Meet

Eligibility under the R&D Tax Incentive 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 s 355-30(2) adds a dominant-purpose bar where an activity (a) is of a kind referred to in s 355-25(2), or (b) produces goods or services, or (c) is directly related to producing goods or services. That is the whole framework; the rest of this article is about the engineering questions it is applied to.

Solution-diffusion transport, concentration polarisation, crystallisation kinetics and colloid stability are settled bodies of principle, so a fouling investigation has something real on which to build a systematic progression — the established-science limb is capable of being satisfied here. The limb that usually bites is the first, because many routine desalination outcomes are determinable in advance: a projection program, a feedwater analysis and an element datasheet will tell a competent process engineer what the plant will do. Difficulty, capital cost and novelty to the company are not the test.

Why Fouling Propensity Is Often Not Predictable from the Published Indices

The industry's standard screening measure is the silt density index, SDI15 — a dead-end filtration test through a 0.45 µm membrane at constant pressure, and the measure feedwater specifications and element warranties are commonly written around. It is a useful acceptance test and a poor predictor, for reasons that are mechanistic rather than procedural:

Scaling sits on firmer ground — saturation indices for calcium carbonate, gypsum, barite, celestite and silica are calculable — but the operative constraint is often not the saturation ratio. It is the induction time before nucleation at the membrane wall, and the extent to which an antiscalant extends it. Antiscalant threshold limits are typically established in controlled synthetic brines; whether they hold in a real concentrate containing competing adsorbents is not something a projection program models.

None of this makes desalination work R&D by default. Sizing a train from a feedwater analysis, selecting pre-treatment from an established framework and setting cleaning intervals from manufacturer guidance are engineering with determinable outcomes. See what does not qualify.

The Concentrate End, and Energy Recovery at Part Load

Recovery ratio governs everything downstream through the concentration factor, 1/(1−R): at 75% recovery every rejected species is concentrated fourfold, at 86% just over sevenfold. The recovery a plant can hold is therefore often constrained by the least soluble sparingly-soluble salt under the specific operating conditions, rather than by the membrane alone, which is what makes concentrate volume reduction an engineering problem rather than a purchasing decision, and the territory of high-recovery staging with interstage precipitation, osmotically assisted reverse osmosis, membrane distillation and electrodialysis, alongside valorisation routes recovering salt, magnesium hydroxide, or acid and base by bipolar-membrane electrodialysis.

At the discharge end the constraints are explicit. The Adelaide Desalination Plant operates under an EPA licence requiring, among other conditions, that salinity in the marine environment be maintained within 1.3 ppt of ambient conditions at 100 m from the diffuser, with continuous monitoring at specified offshore points (EPA SA). Assessing the ecological effect of a hypersaline plume is not a lookup: the ANZ Water Quality Guidelines note that default guideline values are typically derived for marine waters of typical oceanic salinity, that there are currently no default guideline values for estuarine or inland saline waters, and that mixture toxicity at elevated salinity "cannot be readily predicted at present", recommending locally relevant guideline values instead (waterquality.gov.au).

Energy recovery deserves a separate note because its published performance is a design-point figure. An isobaric pressure exchanger is characterised by transfer efficiency, volumetric mixing and lubrication flow, all quoted at rated duty. Mixing matters more than it sounds: concentrate crossing into the feed stream raises membrane feed salinity, which raises osmotic pressure, which raises the pressure the high-pressure pump must deliver for the same flux. At part load, duct velocities and rotor speed fall, mixing becomes a larger fraction of throughput, and the high-pressure and booster pumps move off best efficiency. Australian plants run this way by design: the EPA records that since 2015 the Adelaide plant has operated at reduced capacity, with instantaneous rates of up to 300 ML/day for about 1.5 days per week and maintenance shutdowns in low-demand periods (EPA SA). Where the vendor's curves cover the duty, applying them is engineering; where the turndown sits outside any characterised envelope, an experimental question may exist.

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 brackish groundwater plant abstracts 4.0 ML/d of bore water at 4,200 mg/L TDS, with calcium 480 mg/L, sulfate 1,350 mg/L, dissolved ferrous iron 1.4 mg/L and manganese 0.15 mg/L, at 19–21 °C. Abstraction is held constant throughout, and every recovery figure below is worked against it. Pre-treatment at baseline is cartridge filtration only, so all 4.0 ML/d reaches the RO feed. It runs at 75% recovery, sending 1.0 ML/d of concentrate to a lined evaporation basin already at its consented area. SDI15 on the filtered feed is 2.6 and MFI0.45 is 1.9 s/L² — both well inside specification.

What Is Generally Unlikely to Be Core on Those Facts

Where an RSP Fits

An RSP is a scientific or technical service provider registered in specific research fields — environmental engineering (ANZSRC 4011) among them — that a company can engage to conduct R&D activities on its behalf (business.gov.au). On a membrane plant that work sits upstream of commissioning: framing the unknown, designing the trial matrix and the bench work that locates a mechanism, and specifying instrumentation and normalisation to the resolution the measure actually requires, so the experimental record exists while the work happens (record keeping). There is a threshold point for smaller plants: Qualifying expenditure incurred to a non-associate RSP may still form part of the offset where total notional deductions are below the usual $20,000 threshold, provided the services are within a research field for which the RSP is registered. 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 tiers are covered in our article on the refundable and non-refundable offset; see also claiming R&D under $20,000 and R&D Tax Incentive in Adelaide.

Frequently Asked Questions

Q: Is designing a desalination plant an eligible R&D activity?
A: Generally not, as a core R&D activity. Sizing a train, selecting pre-treatment from an established framework and setting cleaning intervals from manufacturer guidance apply established methods with outcomes a competent process engineer can determine in advance. A core activity may exist in the narrower case where a specific feedwater's fouling or scaling behaviour cannot be predicted by any established index or projection model, and can only be determined by a systematic progression of work conducted to generate new knowledge. You self-assess.

Q: Why does SDI not predict reverse-osmosis fouling?
A: The silt density index is a dead-end filtration test through a 0.45 µm membrane at constant pressure. Material appreciably finer than that cut-off — freshly precipitated ferric hydroxide, for instance — is largely invisible to it; it has no cross-flow term and so does not represent the concentration-polarisation layer that governs deposition at the membrane wall; and it captures nothing of biofouling potential such as available organic carbon or transparent exopolymer particles. A feedwater can therefore return an SDI inside specification and still foul quickly.

Q: Is brine concentrate management eligible for the R&D Tax Incentive?
A: It is assessed as an activity, not as a waste stream. Operating an evaporation basin, or monitoring a discharge against licence conditions, is generally unlikely to be a core R&D activity on those facts, subject to the activity's own facts and the statutory tests — licence monitoring may be an activity associated with complying with statutory requirements or standards, depending on the particular obligation. Determining whether a defined recovery increase can be held on a particular concentrate, where the limiting sparingly-soluble salt's behaviour has no reliable predictive basis, may be a core R&D activity assessed on its own facts.

Q: Can energy recovery device performance at part load be an R&D question?
A: Sometimes. Isobaric pressure exchanger transfer efficiency, volumetric mixing and lubrication flow are published at rated duty, and mixing matters because concentrate crossing into the feed raises membrane feed salinity and therefore the pressure required for a given flux. Where the vendor's characterisation covers the turndown range, applying it is engineering. Where a plant must run well outside any characterised envelope and the specific energy consumption at that duty cannot be determined in advance, an experimental activity may exist — assessed on its own facts against the statutory tests.

Sources & Further Reading

• business.gov.au — Check if you are eligible for the R&DTI

• business.gov.au — Conducting core R&D activities

• business.gov.au — Conducting supporting R&D activities

• business.gov.au — Record keeping for the R&DTI

• business.gov.au — Get help from a research service provider

• legislation.gov.au — Income Tax Assessment Act 1997 — Div 355, incl. ss 355-25 and 355-30

• EPA South Australia — Adelaide Desalination Plant — licence salinity and monitoring conditions

• EPA South Australia — Harmful algal blooms and marine heat waves

• waterquality.gov.au — Accounting for local conditions

• waterquality.gov.au — Toxicant default guideline values for aquatic ecosystems

• Related: environmental engineering research · what does not qualify · what an RSP is · refundable vs non-refundable offset · claiming R&D under $20,000 · more Insights