An Uzbek Refinery Side-Stream at 17000 m3/d: What Coalescing Deoiling Really Proves

Question: I saw that a Chinese company called SINOKLE won a refinery circulating water treatment project in Uzbekistan, with a treatment capacity of 17000 m³/d, using a coalescing pressure deoiler, and with effluent oil and suspended solids both held below 25mg/L. Is there anything technically noteworthy about it? Having worked in industrial water treatment for a few years, here are some thoughts.

The Forgotten Energy Sink

Start with a piece of background that tends to be overlooked. A refinery's circulating cooling water system is a hidden heavy consumer of energy. Many people assume a refinery's energy consumption is dominated by its columns and reactors, but in fact the electricity consumed by the circulating water system accounts for more than 40% of the plant's total auxiliary power consumption. Deterioration of circulating water quality - chiefly the accumulation of oil and suspended solids - directly lowers the heat transfer coefficient of exchangers, forcing the system to increase circulation volume or run at reduced load, which pushes the energy consumption curve upward. So the refining industry takes control of circulating water quality seriously. Taking it seriously does not make it easy.

Three Routes, Three Compromises

Traditionally refineries have taken three routes to remove oil from circulating water.

• Gravity separation (API/CPI oil separation basin). The oil-water mixture flows slowly through a basin and the density difference lets the oil float up. Advantages are simple structure and tolerance of shock loads; drawbacks are an enormous footprint and essentially no effect on emulsified oil or very fine droplets finer than 50μm. Treatment tops out at around 50-100 mg/L and goes no further.
• Flotation (DAF/IAF). Large numbers of microbubbles are injected into the water, the bubbles adhere to oil droplets and rise together, and a scraper removes the scum. It can handle smaller droplets, but it needs a dissolved-air system, a dosing system and a scraping system, so there is more equipment, higher energy consumption and more complex operation and maintenance.
• Chemical demulsification plus coagulation and sedimentation. Demulsifiers and coagulants are dosed to break the emulsified oil apart and then flocculate and settle it. Results can be quite good, but chemicals are a continuous consumable and the cost is not low. More troublesome still, the process generates sludge containing both oil and chemicals, which is hazardous waste, and disposing of it costs money again.

Summed up: either the equipment is complex and expensive to run, or the chemicals cost money and generate hazardous waste, or the precision is insufficient and compliance is hard to reach.

The Fourth Route

The coalescing pressure deoiler that SINOKLE used on this project takes a fourth technical route - purely physical coalescing separation. The core principle is not complicated. Driven by pressure, typically 0.2-0.5 MPa, oily water passes through a coalescing cartridge whose interior is a fibrous medium with a special surface treatment. As small oil droplets travel through the gaps between fibres they are captured on the fibre surfaces and merge. Once countless micron-scale droplets have gathered and grown on the fibres to millimetre scale, buoyancy is sufficient to overcome the resistance of the flow, the droplets detach from the fibres and rise rapidly inside the separation chamber, and are finally discharged from the oil collection zone at the top.

Cartridge selection is the hard part

You cannot simply stuff in a wad of fibre and expect coalescence. Fibre diameter, surface wettability (oleophilic and hydrophobic), porosity and specific surface area all bear directly on coalescing efficiency and pressure drop. On a duty such as circulating water that carries solid suspended matter, the cartridge must also filter - it has to coalesce the oil out and intercept the suspended solids, and it must not clog too quickly. That places considerable demands on cartridge material and structural design.

Integration is where projects are won

How two units run in parallel, how automatic oil discharge is controlled, how the backwash cycle is set, how fluctuations in inlet oil content are handled - all of these surface during on-site commissioning. Delivering stably means the integration design passed the test.

Is This Throughput Actually Big?

Within refinery circulating water side-stream treatment, this throughput qualifies as large-scale plant class. A refinery's circulating water flow is usually on the order of tens of thousands of tonnes per hour, but side-stream treatment generally takes 2%-5%. Converted, 17000 m³/d is roughly 708 m³/h; assuming a side-stream ratio of 3% and working backwards, the corresponding main circulating water flow is around 23000 m³/h - which already represents the circulating water system of a medium-to-large refinery.

Effluent oil and suspended solids both held below 25 mg/L is not a startling figure on its own. Achieving it stably at this throughput is another matter entirely. Once you scale up, uniformity of velocity distribution, cartridge service life and oil discharge cycle all become the points where real projects tend to come unstuck.

Where the Greater Value Lies

This is not a "sold two units" trading story; it is a milestone in technical validation. The Central Asian market - Uzbekistan, Kazakhstan and Turkmenistan - is rich in oil and gas resources and its refining and chemical capacity is far from small, yet environmental infrastructure is generally weak. Many refineries in these countries are still at the stage of separating oil and water by gravity and meeting water quality limits by dilution. If Chinese environmental equipment can prove itself in these markets, the ceiling is high.

This SINOKLE project ran the full chain from design and manufacturing to installation and commissioning, accumulating a set of reusable engineering experience along the Belt and Road route. Operating data at this scale is a technical calling card for entering the Central Asian energy and chemical water treatment market. In the end, when industrial equipment goes global, the most valuable thing is not what it can do, but where it has been done and what the data looks like.