A 10-day sampling study of palm oil mill effluent (POME) shows whether a mill’s effluent is strong and steady enough to justify a biogas plant. In a study of two neighbouring mills in Sarawak, raw effluent at both had a chemical oxygen demand (COD) of tens of thousands of mg/L, which is plenty of energy to recover. The more important finding was that the two mills differed sharply in how steady that strength was.
The challenge: judging biogas potential before committing capital
A plantation group in Malaysia owned two palm oil mills and was considering biogas capture, with the methane either flared to earn carbon credits or used as fuel. Before an investor would consider funding either site, someone had to answer a basic question: how much organic load is actually in the effluent, and does it behave consistently?
Design guides give typical POME strengths, but typical figures do not secure finance. A short, structured sampling programme at each site gives measured numbers that can go into a gas yield model.
The approach: three sampling points, twice a day, for ten days
Where the samples were taken
At each mill, samples came from three points along the existing treatment train, which is a series of open anaerobic ponds followed by an aerobic pond:
- Point 1: raw effluent from the mill, before it enters the ponds.
- Point 2: the outlet of the last active anaerobic pond.
- Point 3: the final discharge to the watercourse.
Samples were taken at 10.00 am and 2.00 pm each day, both within the mill’s 12-hour working day, and sent to a laboratory. Pairing a morning and an afternoon sample tells you something about daily variation. Only pH and COD were reported, using standard APHA methods. BOD, which the report mentions, was not given in the results.
Why three points rather than one
The raw sample (Point 1) tells you the fuel available. The Point 2 sample shows how well the existing ponds already work. The final discharge shows what the receiving environment currently gets. Together they describe the “before” picture, and they give a baseline against which any new digester’s performance can be measured.
What the results showed
| Mill A | Mill B | |
|---|---|---|
| Raw effluent COD (mg/L) | about 29,000 to 121,000 | about 71,000 to 104,000 |
| Last anaerobic pond COD (mg/L) | about 2,200 to 38,000 | about 6,700 to 17,700 |
| Final discharge COD (mg/L) | about 220 to 440 | about 440 to 470 |
| pH, raw / last anaerobic pond / discharge | 4 to 5 / 7 to 8 / 8 to 9 | 4 to 5 / 7 to 8 / 8 to 9 |
Reading the numbers
- Raw effluent is acidic and very strong. A pH of 4 to 5 and COD in the tens of thousands of mg/L is typical of fresh POME. COD is the quantity from which potential methane is estimated, so a high figure means a high energy potential per cubic metre.
- Mill B was consistent. Most readings sat between roughly 85,000 and 100,000 mg/L. That is good news for a digester, which prefers steady load.
- Mill A was erratic. Readings swung by a factor of four, with a single reading above 120,000 mg/L and several below 50,000. Variable strength makes it harder to size equipment, and it increases the case for a buffer or mixing tank ahead of the reactor.
- The ponds already remove most COD. By the last anaerobic pond, COD fell by roughly 85 to 95% at both mills. But open ponds release the methane to atmosphere. The same conversion inside a covered reactor captures it.
- Discharge COD was a few hundred mg/L. That reflects the aerobic polishing stage after the anaerobic ponds.
The inference that the ponds are not working badly, and that gas capture is the opportunity, is exactly why a baseline matters.
Why it matters for the project
The same investor-side analysis modelled both mills with near-identical equipment and cost. The mill expected to yield more carbon credits showed the better return, and the model said that both would be only marginally viable at the carbon credit price assumed at the time. The practical lesson is that returns depend heavily on measured gas potential and on the value of the credits or gas, so sampling data that overstates strength is a risk to the whole financing case.
Where carbon credits are involved, the project must also be registered and validated under the rules of the UNFCCC Clean Development Mechanism, which requires documented baseline data of this kind.
What to consider for your own sampling programme
- Sample over enough days. Ten days showed a four-fold swing at one mill. A single grab sample would have been misleading in either direction.
- Sample the same points at the same times. Consistency is what makes results comparable.
- Cover a representative period. Palm fruit supply varies through the year, so a December study will not capture a peak season. Treat it as a sample, not a guarantee.
- Measure flow as well as strength. COD only becomes a gas yield when multiplied by effluent volume, and this study reported COD and pH only.
- Test more parameters. Add BOD, suspended solids, oil and grease, and temperature if the digester design needs them.
FAQ
What COD does palm oil mill effluent usually have?
Raw POME is commonly in the tens of thousands of mg/L. Here, readings ranged from about 29,000 to 121,000 mg/L across two mills.
Why is raw POME acidic?
Fresh effluent contains organic acids from the milling process. The readings here were pH 4 to 5 before digestion and pH 7 to 8 after the anaerobic ponds.
How long should a POME sampling study run?
Ten days was used here. Longer studies, or repeat studies in different seasons, give a more reliable picture, particularly if load varies.
Do open anaerobic ponds already capture the biogas?
No. They convert the COD but the methane escapes. A covered digester is designed to capture it.
Assessing your effluent?
Talk to us about a sampling programme and we will help you turn the results into a realistic gas yield.