Treating Oil, Gas and Water - more on GOSPs

My apologies that this is running a little late, unfortunately what looked as though they were minor weather problems on national television become a lot more personal when you're waiting for small planes to take you to smaller than normal airports.

At the end of the last tech talk I was commenting on the amount of water that usually comes out of the ground whenever we extract fossil fuels. Once it gets to the surface the questions become two-fold. The first is how do we get it out of the mix, so that we can separate what we would like to have (the fuel/fuels) and then we have the problem of what we do with what is left.

The simple answer (as in easy to write) to the separation of the different components of the fluid is the Gas Oil Separation Plant. However, as you can imagine, if you are tasked with separating, for the sake of example, the liquids and hydrogen sulphide from a gas flow of 1.5 billion cf/day, which is coming from some 87 wells that concurrently produce some 300,000 bd of Arabian Light crude, the actual design of such a plant is anything but trivial. Even the sulfur that is drawn off, at some 90 tons/day must needs be provided for in the design of the plant.

First, as the gas is produced from the wells, it must be collected into a common feed line, through a connector, called a manifold, that takes in the various smaller pipes from the individual wells, and feeds the result out in a larger pipe to the GOSP. Because the pressure that the fluid retains is useful as part of the separation process, we don’t want to lose any more of it than we have to in overcoming the friction in the pipes that it is passing through. (I have a couple of horror stories from my past about occasions when folk who should have known better used pipes that were too small, and ended up reducing both the flow rate and available pressure at the delivery end of the line). The pipes that carry the flow must, therefore, be large enough to carry the flows, though many miles of pipe to the GOSP. Thus we find, at Haradh, that the initial flows were collected into three different manifolds, and that they in turn carried the mix to the GOSP through five pipes which are, depending on flow, either 20 or 30-inches in diameter. (Different pipes are needed to continue to separate the sour (hydrogen sulphide containing) gases from the sweet before it gets to the GOSP.

To construct the plant required

The main plant is made up of 100,000m³ of concrete, 22,000t of structural steel, 410,000 welded joints, 4,000km of cabling, 540km of plant piping, 1,400 items of engineered equipment and 750km of line piping, ranging from 18in to 56in in diameter.

The construction was also expensive of manpower:

For the construction of the facilities, the companies used 32,000m² of office and workshop space, a residence camp for 1,000 men and supporting facilities and a Boeing 737-qualified airstrip, 8,000ft in length with day and night operations. Support systems included 2,500 telephone exchange lines and video conferencing, data networks for 470 users, ground-to-air radio, 306km of fibre-optics and five communications towers.

It cost $2 billion and at peak construction had some 10,600 men working. It took 3 years to build (coming on stream in January 2004).

To sweeten the gas (get rid of the sulfur) it is generally bubbled through columns containing an absorbent liquid (typically sulfur-attracting amines) which remove the sulfur (which can then be recovered by heat in a stripping column, while the amine is then recycled).

In the more recent construction required at the GOSPs put in to add 1.2 mbd of Arabian Light crude from the Khurais addition, which came onstream this year there was an additional consideration. In order to help get the oil out, Aramco is simultaneously injecting 4.5 million bd of treated seawater. The main treatment plant at Qurayyah can process more than 13 million bd. The treatment involves removing particulate solids, ensuring that the oxygen content is below detection (this is needed to prevent corrosion), that there be no scaling products in the water and that there is a minimum amount of microbial content that could lead to biofouling of both the distribution pipeline and the injection wells. (Note that many of these concerns also relate to cleaning up the produced water from the wells before it is re-injected. ) The volumes that are treated are expensive, and thus there has been a recent move to simplify this treatment. Getting the solids out is relatively straightforward. Horizontal sand filters (albeit some 11 ft in diameter, 40 ft long) can treat up to 125,000 barrels of water a day bringing the particulate matter down below 0.2 mg/L. (On a personal note, again, sand filters work very well as long as there is no clay in the water - it takes just a few minutes for clay to coat and plug the top of the filter and make life really, really interesting).

At the same time, to ensure that there is no corrosion in the pipelines that carry the water (perhaps over 150 miles to a well in a journey that might take 36 hours) the pipes themselves are given a special internal coating to try and retain water quality integrity and to ensure that the injection wells do not become plugged. The pumps and equipment are powered by gas turbines.

And speaking of gas the recovery of the oil at Khurais is also expected to generate some 0.3 bcf of a sour natural gas that will be sent to Shedgum to have the sulfur removed, and 70,000 bpd of NGL.

Getting the gas out of the oil (as both Darwinian and idontno commented after my last post appeared on TOD

Very basic, the oil, gas, water, and sometimes sand enter a hydrotreater. The flash gas compressor takes suction from the hydrotreater to remove the gas. The hydrotreater normally will have an electric grid that helps to corral the gas at the top of the hydrotreater. The gas is drawn off and flows to the flash gas compressor. The water and oil separate in the hydrotreater. There’s usually two hydrotreaters and sometimes three to separate the oil from the gas. The velocity of the fluid as it passes through the treaters is critical since if flow is too high there will not be enough time for the oil and water to separate. The water is drawn off for further processing. The sand is separated in the sand separators and bagged for deposit ashore. The oil is processed to get the salt down to specification. Nalco, one manufacturer who fabricates different kinds of production equipment, has a very good web site to visit. They include flow diagrams.

Depending on the gas and oil mix the process of getting the gas out can pass through more than one stage where the pressure in the vessel is dropped, and at the lower pressure the gas bubbles out of the oil. (As noted just as carbon dioxide bubbles out of soda when the can is opened).

In order to accelerate the separation of the water from the oil (remembering that oil floats on water) the vessel is often now heated, so as to speed the process up. Sand in the fluid is going to be an increasing problem as wells get deeper and more of them are horizontal, but it can be exemplified by the way in which the oil and sand are separated up at Fort McMurray, and for that there is a video.

This is part of a series and I am grateful for the help that is given both by those asking questions, and those with practical experience that help with replies.

Produced Water, GOSP's and Saudi Arabia

To the uninitiated the thought of a gas or oil well is one where a pipe goes down into the ground, and out of it flows either a steady stream of oil or natural gas, that is fed straight into a pipeline and then delivered to them (often at what they consider to be an outrageous price) with no further treatment. Or the crude oil that comes out runs straight over to a refinery where (with minimum effort and maximum profit) it is transformed into the gasoline or diesel fuel that they must then again buy at great cost in order to drive in to the liquor store to buy some beer.* The reality of oil and gas production is considerably different, and fluid that comes out of the well is not the ideal that the uninitiated imagines. So today’s topic will deal with the initial separation of a couple of the parts. This is a part of the technical pieces that I write on Sundays about various aspects of fossil fuel production, and it is a relatively simplistic explanation which seems to fit most folks needs, though it also has considerable help from those with more technical knowledge who add comments.

There are three major fluids that come out of a well and these are gas, crude oil and water. If the well is a natural gas one, the oil component will not be the heavier fractions that we associate with an oil well, but rather the higher end liquids such as propane and these are referred to as the natural gas liquids (NGL). NGLs include ethane, propane, butane, iso-butane, and natural gasoline. But today I am going to talk about the water.

Back when I first started writing about this separation oilcanboyd was kind enough to point me to the Produced Water Society.

The Produced Water Society is a collection of engineers and industry professionals with the common purpose to study and improve the separation, treatment, and analysis of Offshore and Onshore Produced Water with the goal to meet the discharge and reinjection requirements of the industry and the environment.

And just to be clear about what Produced Water is:

Produced water is mainly salty water trapped in the reservoir rock and brought up along with oil or gas during production. It can contain very minor amounts of chemicals added downhole during production. These waters exist under high pressures and temperatures, and usually contain oil and metals. . . . . . . The treatment of produced water is a major component of the cost of producing oil and gas. Wells may start out producing little water but sooner or later all oil wells produce a much larger volume of water than oil. The ability to efficiently and economically dispose of this water is critical to the success in the oil production business.

Back in April 2007, oilcanboyd quoted volume flows for the lower 48 US States as being 4.8 mbd of oil, 128 mbd of brine (the typical term for produced water). His number was considerably higher for the water than that offered by the Argonne report given below, though they admit that their count could be significantly under true values.

The changes in pressure, temperature, and the possible access to oxygen when the water reaches the surface, means that the water can precipitate out dissolved minerals and hydrocarbons such as paraffin, which can plug wells that are being used for disposal,

65% of the produced water generated in the US is injected back into the producing formation, 30% into deep saline formations and 5% is discharged to surface waters.

Argonne National Labs recently reviewed the status of this brine, providing not only a review of the process, but also the summary of conditions for each state. They show the relative volumes of water produced, in 2007, by the five largest producing states:

Total produced water generated by wells in the United States in 2007 (with top 5 state producers identified) Source Argonne National Labs

To try and give some sense of the scale of these numbers they point out that Washington DC and its local communities collectively use some 300 million gallons a day, which is only 13% of the amount of produced water that must be dealt with. The water comes from the roughly 1 million oil and natural gas wells that are still producing in the United States. Texas, while the largest producer of natural gas (6.9 tcf in 2007) lagged offshore in the amount of crude that it produced. The national average amount of water produced per barrel of oil was 7.6 barrels of brine, which produced about 87% of all the produced water developed. The average gas well production was around 270 barrels of water per mcf of natural gas. Some 59% of this is reinjected into the producing formations in onshore facilities (only about 9% offshore) in order to enhance production. These relatively large volumes that must be processed and disposed of can control the economics and life of the operation. As the Argonne report notes:

early in the life of an oil well, oil production is high and water production is low. As the production age of the well increases, the oil production decreases and the water production increases. When the cost of managing produced water exceeds the profit from selling oil, production is terminated and the well is closed. This is contrary to the typical production cycle of a coal bed methane (CBM) well. Initially CBM wells produce large volumes of water, which decline over time. Methane production is initially low, increases over time to a peak, and then decreases.

Because the US fields are, in the main, much older from a production point of view, than the average well in the total world, the average water flow is higher, the report estimates that the global value is around 3 barrels of water per barrel of oil.

Because of the high salinity (generally greater than that of seawater) the amount of sodium and salt in the water make it difficult to use for agriculture (which has a very large demand for water in its own right). However in states where the water is reinjected to maintain reservoir pressure even the volumes available may not be enough, and thus one finds, for example in Alaska, that the 842 wells using this EOR used about 1 billion barrels of water in 2007. Given the recent controversy over the disposal of water from the development of the Marcellus shale in New York, it is perhaps interesting to quote the numbers for that state.

The most recent available report is for 2007. According to the 2007 data, 13,113 wells were reported to the division. Of the total, 7,387 were natural gas wells, 4,874 were oil wells, and the remaining wells were gas storage, dry holes, and solution salt wells. The database provided production volumes of 55,001 Mmcf for natural gas and 377,514 bbl for oil. The state-produced water volume was 649,333 bbl from active wells for 2007, which included 215,050 bbl that were associated with water injection wells.

Handling the water from these wells is thus not a small matter, especially in the larger production fields around the world such as Saudi Arabia. When Aramco decide to increase production from a field, or to add another field to their supply network, they cannot just drill another well, hook it into the line and see their exports increase. Because of the nature of the fluid that actually comes out of the hole, it has to be run, first through a Gas Oil Separation Plant or GOSP. Here the oil, formation water and gas that come out of the well together are separated, so that they can be piped to the different treatment plants. (And as a side point readers might want to look at some of the articles on oil production from Saudi Aramco World since they are written more for a family audience than a technical one.) These plants are generally rather large, the one in the article treats 450,000 bd of oil, and they take considerable time to build, install and connect up. Thus when new production is planned one has to wait for the plant to be in operation before the wells themselves can be productive. The new addition at Khurais, for example, required a new central processing plant, and when Haradh Stage 3 began, it had, first to have the new GOSP in place and running, which is was by the second quarter of 2006. Thus the production increments in the country are controlled by the rate at which these can be brought on line. In addition the older ones had to be upgraded, particularly in the controls for the system. (Side comment, though the KSA centralize their GOSPs, they don’t have to be that big. We have had an individual well unit hauled through our yard behind an SUV). But we’ll talk about them a little more another time.

Putting this all together the oil and gas industry have been handling, without significant public complaint, relatively large volumes of water for a considerable time. The processes are handled through the state agencies (which is where Argonne got much of their information) in a set of processes that seem to be under control.

* across the street, last week.