More on refining

Last week I pointed out that the crude oil that comes out of the ground is not made up of a single hydrocarbon, but rather is a mix of different hydrocarbons that have to be separated. And oils from different parts of the world are formed as different combinations of these, and even those from the same country have different properties. Today I would like to go a little further and talk about distillation curves, and because the world supply is changing to heavier crudes then go on to explain a little bit about what is involved in cracking a crude. It is a little more immediately relevant than some posts given that researchers at Purdue have just come out with a suggestion for an improved refining sequence that they claim could improve efficiency by between 6% and 48% depending on how the process sequences in the refinery are re-ordered. And the crude doesn’t have to come out of the ground. NIST have even tested some made from pig manure, with video.

For those new to the site this is where, on weekends, I post a small technical talk, explaining some of the aspects of the fossil fuel business, so as to help understanding of some of the topics on the site. It is a very simple explanation, because of space, and those who wish to ask or expound a bit more are invited to do so through the comments.

When you first get a sample of a new oil, then you will gradually heat it to known temperature levels, generally following a standard method or something more advanced. As the sample is heated, the lighter fractions boil off first, and by plotting the volumes emitted within known temperature ranges, a basic sense of the make-up of that particular oil can be achieved.



Example of two distillation curves from different types of oil

For example, of one looks at the blue line in the above graph, this shows a typical light oil composition, and might well yield, on passing through the first distillation column of a refinery, the sort of separation that I showed in the plot last time.

Distillation products from light oils (Source EIA)

However, as is slowly becoming apparent, we are seeing less of these light oils on the market, and increasingly the world is having to accept a greater proportion of heavier crudes. To understand the difference that this makes, consider the red line in the above graph.

You can see that there is no fraction boiled off until the temperature has passed 250 deg. Which means that, in its untreated state, there is no natural gasoline in the oil. Likely it was too close to the surface and those fractions evaporated away over the millennia (as is the case with the heavy oils in Alberta for eg, though this example is from somewhere else). So how do we change this mix into something more useful?

The answer, not only for this crude, but also for the heavier fractions of the light crude, since in both cases we would like to end up with about 45% gasoline, requires that we crack (or split) the higher carbon molecules into lighter or lower carbon ones. For example say we have a molecule of Cetane (16 carbon and 34 hydrogen atoms or C16H34) if we heat this to a high enough temperature, then we can break it down into some Octane (C8H18), some Hexane (C6H12) and some Ethylene (C2H4). But, obviously, what we would like to do is to control what it is that we break these heavier molecules into, and how much of each we produce.

And so we get into the world of Cracking Oil. After the crude oil has been separated into the different fractions, the gas-oil fraction is then fed to a second heating process, where the fluid is brought back up to a high temperature (perhaps around 735 deg C 1400 deg F) and mixed with a finely powdered catalyst (hence the "cat"). Steam is also added to help with the movement of the mix, and the combined mix is fed up a pipe (called a riser) into a tank and as it flows up and into the tank the gas-oil breaks down into the lower carbon molecules. This happens very quickly, so that by the time the mix is in the tank, the heavy molecules have broken down, and so can be drawn off.

However the catalyst has first to be removed, which happens by passing the flow through cyclones that spins the flow and separates the heavy catalyst, which is collected and fed down a pipe back to be reclaimed and re-used. It needs to be processed since, in part, the process is not perfect and carbon will be deposited on the particles of catalyst, blocking its action. (It can be burned off - providing some of the heat for the process). The catalyst can then be re-used. One operation might use the catalyst at a flow rate of some 55 tons/minute.

The hydrocarbons that flow out of the cracking process are then separated into different fractions, depending on the season more gasoline or distillate might be desired, but the process might yield about 8% coke, 55% gasoline, and 12% light gas oil with about 10% of the flow being a sufficiently heavy product that it is sent back to be run through the process again. Light gas oil becomes diesel and furnace fuel oil. The lighter gas products are usually sent to alkylation, or reforming where the lighter molecules can be combined to give products that can be blended into a gasoline or another desired product.

Different crudes have different mixes of hydrocarbons, and will need different sizes of processing sections that convert these different fractions to usable and desirable products. Refineries have been built to usually handle only a given range of crudes, and thus have to be considerably modified if that mix should change. Sometimes the crude can be even heavier, the sands in Alberta contain an oil which is more conventionally called a bitumen, since it has less volatile material that the heavy oil example I used. That oil must therefore go through an additional step to produce the lighter synthetic oil that can then be distilled and cracked more conventionally. Some of the new processes involved in mining and processing the ore are described in a pdf file by Herron. It also includes the distillation curves for the original crude and that obtained when the THAI process is used.

Distillation curves from the Athabasca Oil Sands (after Herron)

There are also contaminants such as sulfur that can be found in the oil. If the content is higher than 2.5% then the oil is called "sour"; this sulfur must be removed, and while at one time it was used as fuel in the refinery that has been stopped and now it is recovered. That also requires a treatment circuit that must be installed if the mix to the refinery changes.

This is part of an ongoing weekend series on technical aspects of oil and natural gas production, among other things. As ever, if this is not clear, or if there is disagreement then please feel free to post, and I will try and respond.