Pickup Trucks and Oil Refinery Discussion - A Jay and Jimmie Production

Crude Part IIc: Naphtha hydrotreating

At the very top of the fractionating tower all the aromatics/gases – primarily naphtha - naturally rise and collect. At one time, those gases were flared off. Now, they can be captured and used in other parts of the refinery process. Naphtha is also generated by pyrolysis, in the coker, the Fluidic Catalytic Cracker, and VisBreaker units

The first step is naphtha hydrotreating. The hydrotreating is necessary to remove the sulfur and other contaminants to protect the catalysts in the reformer

The naphtha hydrotreater is usually a down-draft design. The gases from the fractionating tower are preheated to about 650 F and up to 12 atmospheres before they enter the downdraft hydrotreater.

The gases first pass through the top-bed catalyst, which is essentially a particulate filter to capture coke precursors and gum. The material can resemble hollow tubes or spheres and has micropores to aid in capturing particulates, and are usually made of a Nickel-Cobalt-Molybdenum in an aluminum matrix. This bed requires regular blowdown to clean away the adhered particulates

Next, the clean gas goes through an arsenic trap catalyst. Depending on crude source, there is naturally occurring arsenic that must be removed to protect human health and the downstream refinery process catalyst materials. The arsenic catalyst is typically a special nickel molybdenum with proprietary ingredients

Next is the actual naphtha hydrotreating bed. This catalyst is typically cobalt molybdenum for straight-run naphtha, but for cracked source naphtha it is usually nickel molybdenum, or a combination bed of both Co-Mo and Ni-Mo. Like all catalysts, it requires regular cleaning and even replacement, typically every 2-5 weeks of constant operation

After the naphtha has been hydrotreated, it is collected and sent to the next stage: Isomerization

As far as the process diagram, sorry the simplest one is still quite complex. If anybody needs anything explained, please let me know

 

Well, at a basic level the crude intended to be used is subject to lab analysis. There are actually lab-scale fractionating towers - about the size of a fridge - and other process equipment that can then do a trial run to empiracally determine lab results

Most of the process changes, such as different pumparound/reflux rates, temps, pressures, are done with the control system. A different target setpoint for temp can be entered and the PID loop will then track the new setpoint

With a modern all-digital process system, such as the SECCO facility near Shanghai that was finished a couple of years ago, these changes can also be done using real-time lab analysis of all the process streams. The system also uses fuzzy logic to better match output to setpoint

If a major change in source crude is necessary, especially if one must transition from Sweet to Sour crude, then the catalyst bed material usually must be completely changed. The catalyst will become poisoned if operated on a different sulfur content

Otherwise the basic design of a refinery remains the same. Most crude source changes can be compensated for with the control system software. Obviously, the newer the plant and process system, the easier to implement that change

When Katrina required a few of the 30-40 year old refineries along the Gulf Coast to shut down, it took up to 3 weeks to reach in-spec. The SECCO facility - thanks to a modern all-digital process control system - went from cold start to in-spec in ONE SHIFT

 

Crude Part III: The FCC (Fluidic Catalytic Cracker)

Moving further along in the refinery stream, the residuum from the bottom of the fractionating tower, and the lowest trays, is processed by the VDU and sent to the FCC.

The FCC is one of the more important process steps in a modern refinery. It’s capable of taking what used to be waste material from the fractionating tower, and turning it into useful products like LPG, gasoline, and naphtha, which can be turned into more gasoline.

You can also derive various process oils that can be further hydrotreated or hydrocracked to yield diesel/jet fuel, in addition to the FCC residuum coke, which can be burned as a fuel within the refinery.

The way the FCC works is a bit hard to understand. The reason why it is called a “fluidic†cracker is that the catalyst is kept in liquid state. There is a vertical column or pipe called a riser, in which the liquid catalyst is kept. The most common catalysts are zeolite and zeolite with additives, such as platinum and aluminum. The catalyst resembles a coarse power, it’s carefully designed as a matrix with a specific surface area and porosity to maximize the cracking process

You need a lot of catalyst to react the feedstock, the typical ratio is 5:1 of catalyst to feedstock. This requires an obvious energy investment, which is mostly recaptured by burning the resulting flue gas from further on the process.

The catalyst is preheated to about 1,300 F and sent to the riser pipe. The feedstock from the VDU is then injected at the bottom of the riser pipe. With the temperatures involved, the catalyst immediately vaporizes the feedstock, catalyzing it into vapors. Usually, in under 30 secs, the vapor has gone up the riser pipe, where it is caught by cyclone cones in a reactor. Think Dyson vacuum

The captured vapors in the reactor are separated from the spent catalyst, which has become coked up with residue. The spent catalyst is sent through a steam stripper to remove any remaining hydrocarbon material, while the captured vapors are condensed and sent to a fractionating column, which functions very similarly to the first stage fractionating tower in the plant.

Once the spent catalyst has gone through the stripper, it is sent to a regenerator. Compressors force in a lot of air to enhance combustion within the regenerator, at temps approaching 2,000 F. The flue gas from the regenerator is sent through an electrostatic precipitator before it is burned as a fuel in other parts of the plant.

The very hot clean catalyst is now sent back to the riser where it again comes in contact with the feedstock from the VDU. This is a continuous process

The captured and condensed vapors from the reactor that are sent to the distillation column encounter pumparounds/reflux similar to the initial stage fractionating tower. Depending on the location of the trays and pumparounds, one can expect heavy oils and slurry near the bottom of the column, lighter oils/gasoline near the middle, and various naphtha’s near the top of the column

The most obvious waste product is all the contaminated water from the steam stripper and also from the distillation column condenser reflux drum. The waste water will need to be treated due to the various heavy metals.

The flue gas is actually burned to heat the regenerator, and to produce electricity for the plant. It must be treated with a scrubber to remove particulate emissions.

 

Depending on the crude source - heavy, light, sweet, sour, etc - and the intended proportion of refinery output (Mostly gasoline, mostly diesel, a mix, etc), the FCC can contribute anywhere from 30-100% of total refinery energy needs.

Indeed, the most modern FCC designs utilize a dedicated Power Recovery Train that are so efficient that the flue gas, and the expansion of the hot flue gas to drive a turbine, will actually generate excess electricity once the coke residuum is burned.

Most oil refineries are completely energy self-sufficient. The proportion of oil consummed in the process of refining a barrel of oil varies depending on the design of the refinery, the newest most modern refineries are the most efficient

For a given barrel of oil, anywhere from 1-6 US gallons is consummed in running the actual refinery

Again, it depends on the design and efficiency of the refinery. In the newest refineries, the ratio of unusable to useful fuel consumption approaches 10:1. Primarily, the coke and heavy sludge that at one time was considered a waste stream is now captured and burned to generate steam and electricity.

An oil refinery has huge demand for onsite steam generation, so it only makes sense to utilize fuels that would otherwise be "wasted." As an example, the membranes that filter desalters, sour water and tray condenser systems has a concentrate hydrocarbon sludge. That sludge can be dewatered and also burned

In almost every case, you would need to ship in coke or use large amounts of wellhead gas - sour gas - or domestic natural gas, in addition to bunker or fuel oil, to initially startup. The refinery is intended to run 24x7, so only in the event of a major maintenance operation, emergency shutdown (Think Katrina), or renovation would extra fuel be needed to startup

 

Re: I have betrayed you all.....sorta

The Bakken deposit in ND and Montana could hold significant reserves, depending on extraction technology, which is maturing

As with any shale deposit, there are significant differences between in-place and technically recoverable.

The oil shale is needed as it contains kerogen, which is a very heavy, thick, tar-like stuff. The shale in northern Alberta will seep in summertime temps, but otherwise is as hard as pavement

You must mine, transport, and pulverize the shale. It must then be heated to upwards of 900 F in controlled conditions to liberate heavy oils, oil vapors, natural gas, and a thick lumpy stuff called char.

In the case of the shale in Alberta, they use the Clark Hot Water Extraction method. This explains the enormous amount of water required per barrel of recovered oil. A conventional oil refinery has about 1:1 ratio of crude to water consumption, tar sands typically 1:8 final crude to water

The mixture of hot water forms an oil slurry that must be separated. The resulting sludge contains a lot of heavy metals that are quite nasty. Recently, Suncor had to issue a public apology over bird deaths after birds drank the toxic tailing pond water

The tar sands in pictures | Greenpeace Canada

The energy consumption to heat the pulverized shale is far in excess of a typical oil refinery running on various liquid crude oils. This explains the relatively low EROEI of shale compared to most liquid crude oils. It is even proposed that on-site nuclear power plants be constructed near Ft McMurray, Alberta, to enhance the EROEI

The oil and gas vapors are captured, condensed, and sent off to a conventional oil refinery for further processing. The synthetic crude is considered "sour" and requires extensive hydrotreating to remove the sulfur and nitrogen

The primary issue with oil shale extraction is the tremendous demand for process water. It's unknown if the water consumption can be maintained

I agree that shale is a possible reserve, but realistically it only breaks even if oil is priced above $70 a barrel

 

For a given barrel of crude oil input, an average of 5 US gallons of crude is needed just to run the process. For sources with extremely poor EROEI - especially oil shale - expect upwards of 15 US gals for every barrel of syncrude

A lot of folks are really getting excited about oil shale, as if it will somehow bring the price of unleaded down to 99 cents a gallon. Never happen, the EROEI is so poor

 

Correct, a barrel is precisely 42 US gallons

That is hard to answer as oil refineries can schedule production for a variety of products to meet expected demand.

For example, say your primary demand is for diesel fuel. You will stage the fractionating tower more to mid-level reflux/pumparound to increase yield of distillates like diesel, kerosene, and jet fuel

The draw from the lower level and the residuum you will also tend to concentrate on increasing draws from vacuum distillation and hydrocracking to increase diesel yield. Ditto for the fluidic catalytic cracker

For refineries in the EU, using Light Brent Sweet, for every barrel of crude they can expect around 25 US gals of low sulfur diesel and 8-10 US gals of gasoline. They can get more gasoline yield by enhancing naphtha hydrotreating and isomerization

Here, with the emphasis on gasoline yields, with typical crude sources - mostly sour - you can expect around 20-25 US gals of gasoline and 8-15 US gals of diesel, again depending on how heavily you hydrotreat and hydrocrack the various residuum and coke flows

A very sour crude source you can expect 10-15% lower yield for highly refined products - diesel and gasoline - but higher yields of coke and process oils

 

I'll try to state it as a percentage, that might be easier to understand

From the CDU:

Residuum: 40%
Gasoline: 15%
Kerosene/Diesel: 10%
Various naphtha's: 30%
Aromatics: 5%

From the CDU the ratios and yields become very fuzzy, again it depends on what the most desired product is. Various hydrotreating and hydrocracking process steps, in addition to vacuum distilation and fluidic cracking and isomerization, can dramatically extend the yield of gasoline and/or diesel

 

Crude Part IV: Isomerization

Once the crude has been vaporized in the CDU tower, the lightest naphtha gases are sent through naphtha hydrotreating, to remove sulfurs and other contaminants, and then sent to the isomerization unit.

Isomerization may seem like magic, but it’s chemical engineering. It’s the process of transforming one molecule into another while keeping the same number of atoms. It depends on catalysts and the application of heat and/or pressure.

The isomerization unit is used to convert the naphtha gases – including paraffin isomers, benzene, and naphthenes - to very high octane blends for gasoline. In essence, it “stretches†the gasoline yield by incorporating what used to be a flare gas that was wasted. The isomerization unit also deals with benzene, a substance that is being regulated in gasoline fuels.

To work, the catalyst material requires feedstock that is completely dry. So the gas from the naphtha hydrotreater must be treated with either a desiccant or dried by other process, or the catalyst will suffer a steep loss of efficiency

For the catalyst, both zeolite and sulfated zirconia have met with success. There are also proprietary catalysts that enhance yield, obviously I cannot discuss the makeup of these catalysts. Chloride injection is also utilized to enhance the catalyst efficiency, and hydrogen is used to minimize coke formation on the catalyst

I’m wondering at this point if anybody is interested in actual chemical equations, or if I should just keep it as simple as possible?

Once the gas has been dried, it is mixed with hydrogen, heated to about 500 F and 100 psi, and sent through a hydrogenation reactor. The hydrogenation reactor will saturate olefins to paraffin and benzene.

The result of the hydrogenation reactor is then sent to the isomerization reactor where the catalysts will work on the benzenes and other gases, to produce the gasoline blends, iso-butanes, and fuel gas for the refinery itself.

This process is difficult to understand without some chemistry knowledge. Rest assured it does work, and the isomerates from the isomerization unit can be further treated by hydrocracking to yield more gasoline blends

Ok, quiz time. In the interest of Fair Disclosure, I have purposely inserted a false statement somewhere above. A pat on the back for the first person to find it

 

Actually, the heavier molecules. Hydrocracking is used on the CDU residuum after the residuum has gone through the VDU. It's also used on the thick slurry from the lowest trays of the CDU

If you wish, I can expound on hydrocracking.

Most of the hydrogen comes from catalytic reforming, which is the next step after isomerization. The aromatics, like butanes and naphthas, will yield the hydrogen once put through the reformer. I will touch on this process step next

You are entirely correct. I cannot mention exact figures, but the petrochemical industry is a *major* consumer of those rare earth metals

 

Ok. Well, if anybody is, please let me know. The material can be difficult to understand

Nope. The last paragraph about isomerization, I stated the next step was hydrocracking. That is incorrect, hydrocracking is primarily used on heavier distillates and residuum, not aromatics

The correct next step after isomerization is Catalytic Reforming. Ok, I promise no more quiz attempts!

Whoops, I should have thought of that. I will modify the diagrams to reflect this. Sorry