Crude oil refining is one of the most fundamental processes in modern industry. Refining crude oil yields not only fuels such as gasoline and diesel, but also raw materials for plastics, synthetic fibers, detergents, packaging, and pharmaceuticals. The first step of crude oil refining is distillation, which boils crude oil to over 350 degrees Celsius and separates components by their differences in boiling point. Although it has been the standard of the refining industry for more than 100 years, it is difficult to avoid enormous energy consumption and carbon emissions.
A research team led by Professor Ko Dong-yeun of the Department of Chemical and Biomolecular Engineering at the Korea Advanced Institute of Science and Technology (KAIST) announced Tuesday that it had succeeded in separating light components from actual crude oil using a cheap polymer membrane. The technology is a next-generation method that simultaneously solves the low processing speed and the costly selective-layer coating problems of existing crude oil membranes, separating crude oil at room temperature without boiling it. The findings were published in the international journal Nature.
"Separation without boiling" matters in refining because of energy use and carbon emissions. The energy used by refineries worldwide for atmospheric and vacuum distillation is estimated at more than 1,100 terawatt-hours (TWh) per year. That is equivalent to the electricity produced by about 130 nuclear reactors of 1 gigawatt capacity running year-round. The carbon dioxide emitted from this also exceeds 160 million tons per year. In particular, the high-temperature thermal energy needed to boil crude oil is so large in scale that it is not easy to replace with electricity, because it requires both facility retrofitting and a large-scale power supply.
For this reason, academia has continued to attempt to change the refining process from "boiling" to "filtering," much like reverse osmosis, which produces fresh water by filtering seawater through a membrane. However, crude oil is not easy to handle with a membrane. It contains thousands of types of hydrocarbons, has high viscosity, and its heavy oil components easily foul the membrane. Until now, academia believed that molecular-level precise separation was only possible by coating the surface with a very thin "selective layer." The selective layer is the key functional layer that actually filters substances, but it is difficult and expensive to manufacture, and defects arise as the area grows larger, limiting its application in industrial settings.
The KAIST research team solved this problem without creating an expensive selective layer, using only a porous polyacrylonitrile (PAN) membrane, which had until now been regarded as a mere support base. PAN is a relatively common polymer material used in acrylic fibers and other products, and the pores of the membrane are larger than crude oil molecules. According to the conventional view, crude oil components should not be filtered properly and should pass through quickly. But in the actual experiments, heavy components in the crude oil attached to the walls of the membrane's fine pores, creating narrow channels of 2 nanometers (nm, one billionth of a meter) or less. The component deposition, which had been regarded only as a fouling phenomenon that degrades membrane performance, instead created precise nano-scale channels. The team explained this as a "self-limiting pore contraction" phenomenon. It means that the pores are not completely blocked, but rather that the crude oil components and the membrane interact to form channels through which only light components can pass.
The experiments also confirmed the performance of rapidly filtering out light components from crude oil. When tested on "Arabian Light" crude oil, the separation speed was more than 23 times faster than the highest level of existing crude oil membranes. In the crude oil that passed through the membrane, the proportion of light components with boiling points below 200 degrees Celsius, such as naphtha and gasoline, rose from 25.1% to 52.0%. Conversely, heavy oil components could not pass through the membrane and were filtered out. The team explained that the separation speed and performance remained stable even when the experiment was run continuously for 28 days.
The team also identified what happened inside the membrane. Analysis showed that long hydrocarbon components with 17 to 33 linked carbon atoms had mainly accumulated inside the membrane. These components settled within the membrane's very small pores, maintaining channels through which only light components could pass. On the surface it appeared that crude oil components had blocked the membrane, but in reality the crude oil and the membrane together created a new filtering screen.
If this technology is commercialized, it could reduce the burden of the distillation process that separates components by boiling crude oil. This is because some light components can be filtered out first by adding a membrane module between existing pipelines, without tearing down the massive distillation towers or halting the entire process. It opens a path to lowering energy consumption and carbon emissions while utilizing existing refining facilities. The team analyzed that, according to process simulation, combining a membrane with the existing distillation process could reduce energy use by 31.6% and carbon dioxide emissions by 37.6%.
Its uses are also diverse. It can be applied throughout refining and petrochemicals, from selecting only the desired components before cracking naphtha, to separating key petrochemical feedstocks such as paraxylene and BTX, and even refining high-grade lubricant base oils. Furthermore, it can be expanded into future industries such as refining pyrolysis oil from waste plastics, recovering solvents used in battery manufacturing, and refining eco-friendly fuels.
However, further verification is needed before it can be applied in actual refineries. "This study is at the stage of confirming performance in the laboratory by cutting small-sized membranes," Professor Ko said. "To use it in industrial settings, we need to verify whether the same performance is maintained even when the membrane area is greatly enlarged, and whether it operates stably over the long term in a module form bundling multiple membranes." Ko also said, "It can also be used for separating naphtha and aromatic compounds, refining pyrolysis oil from waste plastics, and recovering battery materials and solvents," adding, "The key task for commercialization is to scale up the results confirmed in small laboratory membranes into large membrane modules that can be used in actual factories, and to make them capable of running stably over a long period."
