Separate the fractions before selecting treatment
An oil layer separating at rest presents a different problem from a stable emulsion. Identify oil sources, detergents, lubricants, salts, solids and temperature or pH changes. Visual observation can guide sampling, but cannot quantify every fraction.
Removing free oil upstream may avoid unnecessary loading on the membrane stage. Check its effect using actual samples and account for the separated stream. A membrane intended to retain droplets does not automatically remove dissolved organic constituents.
Inopor describes ceramic applications involving emulsions and oil recycling. That experience establishes a technological option, rather than demonstrating flux, service life or permeate quality for a new feed.
Prepare samples representative of the process
Collect water during normal production and events that could limit the installation: product changeovers, washing, salinity increases or detergent discharges. Record the source and time. Do not prepare oil mixed with clean water and present it as equivalent to industrial wastewater.
Use separate samples for analysis and process trials. The laboratory should define containers, preservation and handling for the chosen method. Acidifying an analytical sample and then using it for filtration may change the emulsion being tested. Document any transfer or removal of surface oil.
Record emulsifiers and mixing conditions. Pumps and agitation can change droplets, so describe the sample's route to the membrane. The study by Chen and colleagues uses an emulsion prepared with a particular surfactant and specific conditions; its results should not be presented as plant performance.
Design a staged trial covering complete cycles
Establish reference conditions: composition, temperature, transmembrane pressure, cross-flow velocity or recycle flow, and active area. Select membrane cut-off and geometry around the required water quality, without treating nominal pore size alone as a predictor.
Test controlled changes in pressure and recycle within assembly limits. Then increase concentration to a recovery relevant to the project. A trial keeping the feed tank dilute does not necessarily reproduce the final retentate conditions.
Record flux and pressure over time, feed, permeate and retentate samples, and interruptions. Include several filtration and cleaning sequences. The first hour may show favourable quality while a deposit develops that reduces later production.
Analyse oil and close its mass balance
Define what oil means in the report. EPA Method 1664B measures n-hexane extractable material and, following additional treatment, a non-polar fraction. These are method-defined parameters. They are not automatically interchangeable with any hydrocarbon measurement, turbidity or COD.
Use the same validated method for compared streams and check quantification, blanks and recovery in that matrix. Report permeate below the quantification limit as such, rather than describing absolute absence of oil.
Compare incoming mass with oil in permeate, retentate, upstream separated oil and cleaning liquids. Include inventory change for a batch test. A shortfall may indicate deposits in the equipment or incomplete sampling; it does not demonstrate destruction of the pollutant.
Assess CIP and net production before scale-up
Select cleaning around the deposit and documented limits of the membrane, support, seals and housing. Ceramic resistance does not automatically validate the whole module. Record cleaning concentration, temperature, duration, rinsing and the destination of the CIP liquid.
Compare reference-water permeability before and after the trial under equivalent conditions, alongside oil rejection. Restored flux without retained quality is not successful cleaning. Calculate net production after cleaning, flushing and downtime, and include electricity and residuals management.
Request acceptance criteria and a representative trial as part of a proposal. The useful result combines quality throughout the cycle, stable production and compatible cleaning. None of these can be established from visually clear permeate alone.
Common questions
Is clear permeate free of oil?
Not necessarily. Appearance does not replace a defined oil analysis or identify every dissolved organic constituent.
Should a trial use only the main oil?
No. Detergents, salts, solids and cleaning products can change the emulsion. The decisive trial needs representative water and its variability.