A feasibility study for turning municipal solid waste (MSW) into refuse derived fuel (RDF) has to prove four things: the waste supply, the fuel specification the buyer will accept, the process that gets from one to the other, and the permits and risks around it. A study for an Indonesian landfill showed how each was tested, for a plant taking about 450 tonnes of MSW a day and producing roughly 150 tonnes of RDF.
The challenge: a nearly full landfill and a fuel buyer
The landfill received about 350 tonnes a day (a truck count over a day recorded about 373 tonnes), and the local government wanted less going into it. Nearby factories burned coal in their boilers and could use a cheaper supplementary fuel. The idea was to sell RDF at the plant gate to two or three of them, while also mining the old waste to reduce the landfill.
Test 1: the waste
A waste composition analysis found food waste made up around 64% of the blended input and plastic about 29%, with an average moisture content of about 46%. The site held roughly 600,000 m³ of old waste. At an assumed density of 300 kg/m³ that is about 185,000 tonnes, to be mined at roughly 100 tonnes a day over five to six years and blended with fresh waste at about 350 to 100.
The study noted that the collection area covered only about a quarter of the area it should, so waste volumes were likely to grow, which is the reason for sizing the plant at 450 tonnes a day.
Test 2: the fuel specification
| Property | Buyer’s requirement |
|---|---|
| Moisture | Under 20% |
| Chlorides | Under 0.5% |
| Sulphur | Under 1% |
| Calorific value (as received) | About 5,000 kcal/kg |
| Particle size | Under 200 mm |
For comparison, the incoming waste had a calorific value of about 2,340 kcal/kg, and the design RDF about 5,030 kcal/kg, close to a mid-range coal. Moisture dropped from about 45% to about 16%.
Why 200 mm and not 50 mm
A particle size of 200 mm needs only a primary shredder. Going to 50 mm needs a secondary shredder, which raises capital and running costs, wears out faster, and copes badly with the soil and organics in landfill-mined waste. Choosing 200 mm avoided that.
Test 3: the process
About 38% of the incoming waste by weight ended up as RDF. The line was:
- Hopper and bag opener to open bags.
- Picking line to take out glass, stone and hazardous items by hand.
- Rotating trommel, which removes fine, wet food waste (the largest single removal).
- Second picking line and overhead magnet for ferrous metals.
- Spinner, a hybrid air classifier and shredder that cuts to under 200 mm and drops heavy items out.
- Shredder, then a rotary-drum dryer used only when the waste is wetter than expected, for example in the rainy season.
Operation was planned for 15 hours a day with two hours reserved for maintenance. Baling was left out because it is only worthwhile over around 200 km, and the buyers were about 20 km away. Two buildings were planned: about 1,200 m² for processing and about 3,000 m² for RDF storage, enough for roughly two weeks while a buyer’s boiler is shut for maintenance.
Test 4: permits, contracts and risks
Because the plant would sit inside an existing landfill, a simple environmental management and monitoring statement was expected rather than a full impact assessment. The contracts were a joint development agreement with the local government, a waste supply and build-own-operate-transfer agreement, and a fuel supply agreement with take-or-pay terms. The study scored risks by severity and likelihood. The highest were:
- RDF quality, because a buyer can reject off-specification fuel at the supplier’s cost. Mitigation: regular testing and conservative design.
- Price negotiation and reliability of the plant.
- Local opposition along truck routes. Mitigation: scheduling, routing and aerobic stabilisation to reduce odour.
- Exchange-rate risk, handled by linking price to coal.
Build time was estimated at about eight months from detailed design, with civil works in the heavy rains (May to September) the main threat to the programme. The plant was expected to create 40 to 45 local jobs, with scavengers already on the site employed as sorters.
What to consider for your own project
- Start with a buyer. Specification and price come from the buyer, not the plant.
- Measure the waste. Truck counts and composition studies replace council estimates.
- Be conservative on yield. The desk calculation gave more RDF than the model assumed.
- Protect against rain. Cover processing and storage.
- Agree rejection terms in advance. They decide who carries quality risk.
For the UK picture, see Defra’s waste and recycling statistics. A related post covers landfill mining to energy.
FAQ
What is RDF?
Refuse derived fuel: processed waste, shredded and dried to a fuel specification.
How much RDF comes from a tonne of MSW?
Here, about 38% by weight, with a conservative 30 to 35% used for planning.
Why blend landfill-mined and fresh waste?
It reduces the landfill over time while keeping plant throughput up.
Considering an RDF project?
Tell us about your waste and potential fuel buyers and we will outline the checks.