Separate three questions before testing
The first question is which fraction is physically retained. The second is how much carbon remains in the filtrate. The third is whether the remaining material can be biologically consumed under the available conditions. Treatment selection depends on which of these limits the intended use. The IntegraTec manual distinguishes particle removal from dissolved-substance removal; its behaviour should not be extrapolated into a universal percentage for every UF system.
Report complete units. COD is expressed as mg O2/L and TOC as mg C/L: they represent different balances. COD here means chemical oxygen demand, while DOC means dissolved organic carbon. Distinguishing the measures matters when laboratories and equipment suppliers use different terminology.
| Measure | Contribution to diagnosis | What it does not demonstrate |
|---|---|---|
| Total COD | Chemical oxygen demand of the whole sample | Biodegradability or molecular size |
| Filtered COD | Fraction passing the stated analytical filter | Exact passage through the UF membrane |
| TOC and DOC | Total carbon and the filtered fraction | Toxicity, chemical identity or biodegradability |
| BOD or respirometry | Biological response under defined conditions | UF performance or absence of inhibition under other conditions |
The soluble fraction depends on the procedure
USGS explains that the distinction between dissolved and particulate carbon is operational and based on filtration; colloids can overlap this classification. EPA's nutrient manual notes that COD filtered through 0.45 µm can include fine colloids, and distinguishes that measurement from flocculated and filtered COD. A result described as soluble therefore needs its filter and procedure stated.
Do not compare filtered feed COD with total outlet COD as though they were the same fraction. Request paired UF feed and filtrate samples and measure total and filtered COD in both, alongside TOC and DOC where useful for the decision. Record any upstream coagulation, adsorption or oxidation: the result belongs to the tested treatment combination, rather than to the module alone.
Design a test that addresses the intended use
Define the acceptable maximum for the relevant product-water variable and the feed range the test must cover in advance. Sample stable operation and episodes representing production changes. Retain blanks, duplicates and quantification limits; differences small relative to method uncertainty do not demonstrate meaningful removal.
When investigating biodegradability, the laboratory should define inoculum, temperature, duration and controls for activity and inhibition. EPA describes COD fractionation and the use of respirometric techniques when estimation assumptions do not hold. An isolated BOD/COD ratio provides an indication; it is not universal proof of industrial effluent behaviour and does not by itself establish the need for biological treatment.
Interpret the result without assigning unmeasured mechanisms
If total COD falls while filtered COD changes little, the observation is consistent with retention of a particulate fraction. If DOC also decreases, investigate the contribution of upstream treatment and whether the effect persists over several cycles; an initial decrease may not represent steady operation. If biodegradable load remains in the filtrate, a biological stage may warrant a specific trial. If a refractory fraction persists, compare alternatives targeted at the compound or fraction that limits the use.
Also check matrix-related analytical interference. EPA Method 410.4 identifies chloride as a positive interference in its COD determination. A useful buyer's report includes original data, uncertainty, procedures, conditions and conclusions limited to what was observed. This allows the organic quality of the product to be specified separately from its optical quality.