Enhancing recovery rates with polymer

Norwegian oil-fueled welfare might be extended for many years to come by the use of very long water-soluble polymers. However, polymers are prone to succumb to several stresses at the same time as having uncertain environmental effects.

Three researchers. Three scientists in white coats. PhD students Irene Ringen and Eystein Opsahl are together with post-doctoral researcher Dmitry Shogin studying how polymers can increase oil extraction on the Norwegian continental shelf.

Nonetheless, engineering and research efforts at the National IOR centre are working towards designing the ultimate formulation that can bring technical and environmental performance to acceptable levels. In this text we will try to explain in layman’s terms what this means.

Injecting water to boost oil recovery rates is a well-established method in the industry. Water is injected into the reservoirs as the reservoir pressure drops below ambient pressure so that the producing flow won’t stop. This is known as water flooding or secondary oil recovery.

But, water flooding has its limitations, since its viscosity is low (or ability to flow is high), it follows the path of least resistance through the reservoir and flowing right past the oil-bearing rock layers. This is also known as “viscous fingering” for those of you who like to dig a little deeper.

Scientists are therefore working on flooding solutions with higher viscosity than water. One way of increasing waters viscosity is to dissolve polymers that renders the solution thicker, flowing almost like syrup. This in turn creates a more uniform resistance within the reservoir when injected, flowing through the reservoir in a fan like manner getting to the oil off the beaten track. This is known as polymer flooding and is a sub-discipline of enhanced oil recovery (EOR) or tertiary oil recovery.

Now, what is a polymer? Overwhelmingly many familiar compounds are in fact polymers per definition. One example is our DNA, proteins are another (found for example in egg white), and our blood is teeming with it in the shape of various proteins, polysaccharides and whatnot. A potato is largely a giant reservoir of the polymer starch. And as you might know, dissolving starch in water makes it very thick indeed, and this is exactly what we are doing. However, bacteria tend to rather enjoy starch a little bit too much and this is a huge problem within the reservoir at the same time as it is a blessing for the environment. These are all known as biopolymers, and could in theory all work for polymer flooding purposes. On the other hand, we want to use synthetic polymers that have some practical benefits albeit the environmental part is, let’s say blurry in a lack of better words. Synthetic polymers are polymers which are not found naturally and is made by humans, much like plastics you encounter everyday. The synthetic polymers we are talking about are already being used widely in medicine, agriculture and waste-water treatment and they are mostly based on variants of polyacrylamide.

Anyhow, since polymer solution behaves very differently from regular water, you cannot simply pump it into the reservoir as before. It is also very easy to make costly mistakes. Understanding the properties of polymers are essential for the scientists who are working on this.

Dmitry Shogin is a physicist and mathematician at the University of Stavanger who is associated with the research project. He is studying how the polymers behave on a microscopic level. This is based on a simple fact.[EO1] 

“The engineers observe many interesting effects regarding polymers that they are not able to explain. Not being able to explain these effects means not being able to predict them, which is what you want”, explains Shogin.

Shogin refers to two balls that are connected by an elastic spring, a simple molecular model that can be applied to all types of polymers.

It is important for Shogin to see how viscous fluids behave when they are subjected to external stress. Here we distinguish between Newtonian and non-Newtonian fluids. Water for example is a Newtonian fluid. If you stir sugar in a cup of tea, the tea will move towards the sides.

“Polymers are non-Newtonian fluids and will therefore behave quite differently”, says Shogin.

Imagine whisking egg whites. The fluid will not move towards the sides, as tea does, but will move up the beaters.

Dynamic polymers

Then you have the dynamic properties of the polymer. If you stretch a spring, it is typically linear (the spring force increases proportional to the stretch. This does not apply to polymer molecules! You can only stretch them to a certain length, before they break and the polymer will start losing its properties.

“How fast you reach this limit is one of the parameters that is incorporated into my model”, says Shogin.

Furthermore, the configuration of the polymer molecule are continuously being changed in the flooding solution, and one of the model parameters describes how fast these changes can occur. Polymers behave differently in channels with different geometry. Small individual pores in a reservoir are an example of a quite complex geometry.

Shogin aims to develop a model that enables scientists to predict what happens in more complicated situations, such as in oil reservoirs, and then control these results experimentally.

Sandbox lab experiments

The challenge is to see what happens with the polymer if it is degraded and loses its properties. How do the scientists work to find the polymer that eventually can be used in flooding solutions?

Irene Ringen has a master’s degree in petroleum engineering from the University of Stavanger. She is one of the PhD students at the National IOR Centre of Norway. Mathematician Oddbjørn Nødland is also part of the PhD team and will be applying mathematics to practical problems.

“We aim to achieve a more stable oil displacement with the polymer solution, as it is more viscous than water”, they explain. Water has a tendency to flow past the oil and is therefore not an effective solution”, says Ringen.

She is one year into her research project and is currently testing how different polymer solutions behave when they are transported through sandstone[EO2] . For example, various salts are added to the polymer to see if and how its behaviour changes. Solutions with lower salt concentrations are known to be beneficial for water flooding, and so far the results show that the polymer becomes more viscous if you remove salt.

Affected by stress

The results from the experiments form the background for simulations performed by Oddbjørn Nødland. These simulations are being performed using IORCoreSim, an EOR (Enhanced Oil Recovery) simulation tool which are under continuous development at IRIS (International Research Institute of Stavanger).

“Human beings are affected by stress and so are polymers.” Stress is defined as force per unit area and comes in various forms, anything that can impart energy to the system is stress. In this setting shear forces are of most interest (shear rate basically describes how violent the flow is). The dependency of the polymer viscosity on applied stress is complex, but it is often useful to think of the viscosity as a function of shear rate, the rate of change of the fluid velocity across layers in the fluid. 

One effect Nødland is considering is shear thinning, which means that the higher the shear rate, the lower is the viscosity becomes.

“However, when you reach a critical rate, the viscosity can suddenly start to increase again, an effect which is referred to as shear thickening”, says Nødland. Both shear thinning and shear thickening flow effects are included in the numerical model in IORCoreSim.

Another aspect of the model is mechanical degradation. This is when molecules are subjected to such strong forces that they are physically torn apart, for example when they leave the injection well starts flowing into the reservoir.

So far the model has been a success.

“It matches our observations; the measured pressure drops, and the average alterations in the polymer properties, as obtained from the change in viscosity”, says Nødland.

In parallel with Ringen and Nødland’s work in the lab, environmental toxicologist Eystein Opsahl is studying the long-term effects of EOR polymers on the marine environment and how they affect life there.

The “28-day degradation test” stick in the wheel

These polymers that might end up in the sea after use The environmental fate is of extreme interest. According to the activities regulation §63, any chemical must undergo at least 60 percent degradation within 28 days, measured as respiration. This percentage translates to the amount of substance generic oceanic bacteria are able to devour and respire.

“However, the sheer size of the synthetic polymer molecules is a physical obstacle for the bacteria. They are simply too large to be eaten”, says Opsahl.

Then why not use biopolymers that bacteria can digest due to specialized enzymes? “Everything  comes at a price. Since bacteria can degrade biopolymers so extraordinarily fast, you lose the important viscosity and may even clog the reservoir if you don’t control them with ample use (read vast amounts) of undesirable biocides ”, explains Opsahl. “We just have to work out how to overcome this paradox”. Especially considering the regulations specifying that the degradation needs to be facilitated biologically. On the bright side, the polymers seems to be very benign and not toxic at all, as far as we know, but we will find out for sure.

Withal, it is far from hopeless. Synthetic polymers also degrade, but the bacteria are left on the sideline. We talked about stresses earlier, and this is what ultimately degrades them. It is just a matter of time and place. However, we do not know which stresses contribute the most how fast it happens. If we can find this out, then we may have some good arguments as to why these polymers are what we should pursue.

What happens is: imagine a mile long spaghetti rolled up in a coil, this is what a polymer looks like. It would be too big to fit in your mouth to eat, just like bacteria experiences it. Then you cut the spaghetti in the middle. You still have two very large coils of spaghetti. For the biopolymer case, the bacteria are equipped with scissors able to cut the spaghetti ad libitum. For the synthetic, they are left with only their mouth.  So the “scissor” in the latter case is left to the circumstances, stresses being either UV-light, shear, temperature or oxidation with more function as scissors here. Eventually as the spaghettis are cut again and again the bits eventually becomes, small enough to be eaten, in theory. However, this scission is very hard to measure as opposed to measuring the respiration after the bits are eaten.  

Eystein Opsahl is benefitting from the remarkable technological development we have seen since the beginning of EOR in1970s, especially within data processing and electromagnetic theory. He ages the polymer in simulated marine conditions for longer periods of time and measures changes in the molecular weight distribution, the amount of scission, using multi angle laser light scattering. Using a few devices and computer programs devices each worth a couple of million kroner, Eystein Opsahl aims to find a theory for polymer degradation relating structure with environmental fate.

This will be a big step towards the overarching goal of the National IOR Centre of Norway: recovering more of the residual oil and thereby providing more welfare to the Norwegian people in a sustainable manner.

Text: Elisabeth Hovland