Write a constituent-based specification
State hardness, sulphate, sodium, chloride and process-relevant compounds, alongside conductivity and required flow. A hardness-scaling problem may need a different treatment from sodium reduction for an industrial recipe. The water-quality duty should guide selection rather than the commercial name of the technology.
EPA distinguishes common NF uses, including hardness and certain organic-precursor reduction, from RO’s broader desalination role. These are product families containing different models. NF elements do not all share one selectivity, and RO elements do not all deliver identical capacity or perform equally under the same conditions.
| Requirement | Options to compare | Verify |
|---|---|---|
| Reduce hardness while retaining some salts | NF and alternatives | Target-ion passage |
| Broad salinity reduction | RO and alternatives | Final quality and energy |
| Separate organics and salts | Specific membrane selection | Trial with actual mixture |
Do not convert MWCO into an ionic guarantee
Molecular-weight cut-off characterises a test; it does not independently establish the passage of every salt. NF performance also involves membrane charge, pH, composition, concentration and hydraulics. Strong sulphate rejection does not automatically establish equivalent nitrate or chloride rejection.
Request the test conditions and a projection using complete water analysis. For complex organic compounds or process mixtures, consider a specific separation trial. Conductivity may look satisfactory while a constituent making only a small contribution to the overall reading still falls outside the intended-use target.
Compare recovery and quality on the same balance
Use the same usable-water flow, temperature and feed quality when comparing NF and RO. Review recovery, pressure, element count and any need for blending or a second treatment step. Lower pressure does not guarantee lower cost per cubic metre if another separation is needed to meet quality.
Review concentrate chemistry too. Selective retention of calcium and sulphate changes its composition and precipitation risk. The design must account for the individual ions rather than applying one concentration factor indiscriminately to species that exhibit different membrane passage.
Define a trial that can resolve the choice
Set acceptance criteria for target constituents, net output, consumption and stability. Collect feed, permeate and concentrate during the same period, documenting hours and conditions. Variable water needs a representative campaign rather than relying solely on a prepared laboratory sample.
Require quotations to distinguish equipment, pretreatment, dosing and post-treatment. The appropriate option delivers the required water with a defensible operating envelope. Where information remains incomplete, a well-defined test is more useful than comparing datasheet rejection percentages obtained with different test waters and operating conditions.
Common questions
Is NF always simply low-pressure RO?
No. Selectivity can be the decisive difference. An NF membrane may allow passage of salts that the application specifically needs to reduce.
Can conductivity and hardness alone select the process?
They provide an initial indication, but a reliable projection needs ionic composition and conditions. Identify the constituents that limit the intended use as well.