Lubinpla

How to run an RO antiscalant trial with normalized flow

Summary

A reverse osmosis (RO) antiscalant trial takes 5 steps: record a baseline, normalize flow and pressure drop, step the dose, apply stop rules, autopsy if needed. The steps apply to brackish water and groundwater RO systems where scaling sets the recovery limit. Write the stop rules before the first dose, so the results cannot move them.

  • ASTM D4516-19a standardizes RO permeate flow and salt passage to constant reference conditions so data taken at different pressure, temperature, conversion and feed concentration can be compared (ASTM, 2019).

  • The FilmTec technical manual recommends cleaning when normalized permeate flow drops 10%, normalized salt passage rises 5 to 10% or normalized pressure drop rises 10 to 15% (DuPont, 2026).

  • A feed temperature drop of 4 °C causes a permeate flow decrease of about 10% in an RO system, which is normal and not fouling (DuPont, 2026).

  • In a groundwater RO pilot, five antiscalants at supplier-recommended doses of 2.5 to 5.0 mg/L could not stop last-stage permeability decline at 85% recovery (Mangal et al., 2022).

This method is for a field trial on an operating RO plant or a pilot fed with the same water. It does not cover seawater systems run near their osmotic pressure limit, membrane cleaning chemistry, or antiscalant approval tests that a membrane maker runs in the laboratory.

What do you need before starting an RO antiscalant trial?

Before the trial starts, you need a full feed water analysis, the planned recovery, a scaling projection for the concentrate, calibrated instruments, a normalization method (a way to convert daily readings to fixed reference conditions), and the antiscalant supplier's dose recommendation and safety data sheet (SDS).

The feed analysis should include calcium, magnesium, barium, strontium, bicarbonate, sulfate, phosphate, silica, iron, aluminum, pH and temperature, because each one can set a different scaling limit. Calcium carbonate scaling of the concentrate is calculated with ASTM D3739-19, the Langelier saturation index, for concentrate between 10 and 10 000 mg/L total dissolved solids (ASTM, 2019), and with ASTM D4582-23, the Stiff and Davis stability index, above 10 000 mg/L (ASTM, 2023).

Supplier projections can disagree. In one groundwater pilot, seven antiscalant suppliers' projection programs gave maximum recoveries from 77 to 89% for the same feed water and named different limiting scalants (Mangal et al., 2022). So treat the projection as a hypothesis that the trial tests.

Keep trial records to a recordkeeping guide such as ASTM D4472-23, which covers RO and nanofiltration operating, pretreatment and maintenance logs and states that it does not replace the plant's operating manual (ASTM, 2023). Store and dilute the antiscalant as its SDS directs.

How do you run an RO antiscalant trial, step by step?

Each of the five steps has a number fixed before the trial starts. Dose and duration figures below are illustrative unless a source is given; take the actual values from the antiscalant supplier and the membrane maker.

  1. Record a baseline. Run the RO at the current treatment and the target recovery, and log the data normalization needs every day; DuPont recommends daily normalized data for early detection of scaling (DuPont, 2026).

  2. Normalize flow and pressure drop. Convert each stage's permeate flow, salt passage and differential pressure (the pressure drop across the stage) to reference conditions with ASTM D4516-19a (ASTM, 2019) or the membrane maker's program.

  3. Step the dose. Start at the supplier's recommended dose and hold each dose for a fixed period, for example 14 days (illustrative), before changing it.

  4. Apply stop rules. Compare each day's normalized values with the baseline; when a written limit is crossed, stop, step back or clean.

  5. Autopsy if needed. If the tail element (the last element of the last stage) loses performance, send it for membrane autopsy, a destructive analysis that identifies the deposit.

Change only one thing at a time. Start a recovery increase and a new antiscalant on the same day, and there is no way to tell which one caused the change.

How long should the RO baseline run before the new antiscalant is dosed?

Run the baseline until normalized permeate flow, salt passage and stage differential pressure are stable at the trial recovery, and long enough to cover the feed temperature and flow changes the plant normally sees. No standard sets a field baseline length. For the laboratory antiscalant compatibility test, the FilmTec manual sets a baseline of at least 12 hours without chemical addition, followed by at least 1,000 hours of continuous antiscalant dosing (DuPont, 2026). A field baseline usually needs longer, because a scaling trend in a tail element can take days to show.

The groundwater pilot reported by Mangal and co-authors ran 10 days at 80% recovery and 32 days at 85% recovery without antiscalant; the antiscalant runs, in a second vessel configuration after a 5-day run without antiscalant at 85% recovery, lasted 3 to 5 days each (Mangal et al., 2022). Those periods served a study, not a plant trial, but they show the time scale: without antiscalant, the runs were long enough to see last-stage permeability fall at 85% recovery and hold constant at 80%.

As a practical rule, make the baseline at least as long as one dose step, so the two periods compare on equal terms. Write its length into the trial plan before the trial starts.

How do you normalize RO permeate flow and differential pressure for a trial?

Normalizing converts each day's readings to the same reference temperature, pressure, recovery and feed concentration, so a change in the normalized value reflects the membrane, not the operating conditions. ASTM D4516-19a is the standard practice for this: it standardizes permeate flow and salt passage for spiral wound and hollow fiber systems, and for a brine-staged system it standardizes each stage separately (ASTM, 2019). The practice applies to brackish waters and seawaters and is not necessarily applicable to waste waters (ASTM, 2019).

Temperature is where normalization matters most. A feed temperature drop of 4 °C causes a permeate flow decrease of about 10% (DuPont, 2026), as large as a typical cleaning trigger. Without normalization, a cold week during the trial looks like scaling, and a warm week hides it.

Take the start of the baseline as the reference, because normalization against initial performance shows changes between day one and the current date (DuPont, 2026).

Record differential pressure across each stage, not only across the whole array, since scale usually forms first in the last stage, where the concentrate is most concentrated (DuPont, 2026).

What stop rules should end an RO antiscalant trial early?

Stop the trial early when normalized permeate flow, normalized salt passage or normalized stage differential pressure crosses a limit written before dosing. The membrane maker's cleaning triggers are the outer limits. The FilmTec manual recommends cleaning when normalized permeate flow drops 10%, normalized salt passage increases 5 to 10%, or normalized pressure drop increases 10 to 15% (DuPont, 2026). Its cleaning section adds that cleaning is recommended when the pressure drop increases 15% (DuPont, 2026). Use the manual for your own membrane; other makers may set different values.

Set the trial limits inside those triggers so the trial stops before the membrane needs cleaning. A step-back limit at half the cleaning trigger, such as a 5% drop in normalized permeate flow in the last stage, leaves time to react (illustrative). Each limit gets three written outcomes: continue, step back to the previous dose, or stop and clean.

Write a non-membrane stop as well: a pretreatment upset, a recovery change or an instrument failure ends the dose step, because the manual names pretreatment, pressure control and recovery problems as causes to check first (DuPont, 2026). For the general method of fixing limits before results arrive, see how to set success and stop criteria for a chemical trial.

How do you read RO antiscalant trial results?

Compare each dose step's normalized values with the baseline at the same recovery, stage by stage. A result counts only if the change is larger than the baseline's day-to-day scatter.

The check table below decides whether to continue, step back the dose, or stop and clean. Replace limits marked illustrative with values from your membrane maker's manual.

Check

Pass

Fail and what to do

Normalized permeate flow vs baseline

Within baseline scatter

Drop of 10%: clean (DuPont, 2026); smaller steady drop: step back the dose

Normalized differential pressure per stage

Flat in every stage

Rise of 10 to 15%: clean (DuPont, 2026); rise in last stage only: suspect scale

Normalized salt passage

Within baseline scatter

Rise of 5 to 10%: clean (DuPont, 2026)

Scaling index at trial recovery

Within supplier's stated limit for the dose

Above it: lower recovery or change product (illustrative)

Tail element condition

No weight gain or deposit

Heavy or coated: autopsy

Of these rows, stage differential pressure most often separates scale from other fouling. A rise in the last stage alone, with the first stage flat, usually points to scale rather than particle fouling, which tends to show in the first stage (DuPont, 2026).

In the groundwater pilot, permeability fell in the last stage at 85% recovery with all five antiscalants, while it stayed constant at 80% recovery without antiscalant (Mangal et al., 2022). There, recovery, not dose, set the limit.

Report each dose step as one short line per stage: dose, days held, normalized flow, salt passage and differential pressure against baseline, and the decision taken.

When does an RO antiscalant trial need a membrane autopsy?

Send for a membrane autopsy when the tail element loses performance and the trial data cannot tell which deposit caused it, because the next decision depends on which scalant it is. The FilmTec manual's troubleshooting section says scaled elements are heavy and recommends autopsy of the tail element to analyze the membrane for scaling, with the scale type identified by chemical analysis, EDXRF or ICP analysis (DuPont, 2026). In the groundwater pilot, a permeability drop of more than 20% in the last stage triggered autopsy of the tail element (Mangal et al., 2022).

In that study, the autopsy changed the conclusion. Supplier projections had pointed to calcium carbonate or calcium phosphate as the limit; the analysis found that "calcium phosphate was the primary scalant causing permeability decline at 85% recovery and limiting RO recovery" (Mangal et al., 2022).

Membrane autopsy destroys the element, so plan a spare and agree who performs it: the membrane maker or a specialist laboratory. If the deposit type is unknown, call a membrane specialist before cleaning.

Lubinpla's trial preparation task starts from the open checks left in a product recommendation the customer accepted, and it fixes the criteria and analysis method before any results come in.

Frequently asked questions

What antiscalant dose steps should an RO trial use?

Start at the antiscalant supplier's recommended dose for the feed and recovery, then step down or up one change at a time, holding each step for the same fixed period. In one pilot the supplier-recommended doses were 2.5 to 5.0 mg/L (Mangal et al., 2022). Avoid overdosing, which the FilmTec manual warns against (DuPont, 2026).

Can an antiscalant trial raise RO recovery?

Only if the scalant that limits recovery is one the antiscalant controls. In a groundwater pilot, antiscalants were unable to prevent calcium phosphate scaling or to achieve 85% recovery (Mangal et al., 2022). Identify the limiting scalant before you plan the recovery step.

Should the trial line run beside an untreated line?

A parallel reference line helps when the feed changes. The FilmTec manual describes laboratory setups with a standard line and a test line fed the same water, so an effect can be separated from feed changes (DuPont, 2026). On a single-train plant, the baseline period plays that role.

Related posts