Me, This Week in Energy, and why Halliburton might refuse EPA

This afternoon I was an invited guest on “This Week in Energy” with Nikki Gordon-Bloomfield and Bob Tregelus. Among other things we talked about the fracking process that is being used to help produce the natural gas from shales such as the Marcellus and Haynesville. In the course of the discussion I was asked why of the nine fracking companies that EPA asked for their formulae, only Halliburton had refused the request. Bob pointed out that they were going to be subpoenaed and thus would have to give up the information anyway. I have discussed some of the problems of stimulating a well with hydraulic fracturing, both real and less so, on this site last March when the public perception of the technology began to change.

In the possible explanation I am going to give, you need to know that this is purely a supposition and the chemicals that I am going to mention are put forward out of my own head, as it were. I have no real clue as to why Halliburton are acting the way that they are, and am only building a hypothesis that might only through some slight possibility have any approximate relation to the truth.


In the evolution of the technology that has made production of the gas shales possible several different technologies had to be developed. The rock (which is actually a mudstone) has a very poor natural permeability. I.e. it is very difficult for fluid to flow through the rock, because the passage ways are very narrow and not very well connected. Thus the normal vertical wells would not produce very much oil or gas when drilled through the shale reservoir, certainly not enough to be profitable. The first beneficial development was, therefore, the ability to drill horizontally after the well had reached the reservoir depth. Once this was possible, then the length of the well that was exposed to the reservoir (which might be only 30 ft thick) would increase from that 30 ft to perhaps 10,000 ft. Since the amount of fluid flowing into the well is a function of the length of the exposed well in the rock, when the reservoir rock has a normal permeability this is enough to increase production significantly (as for example in the new wells in Saudi Arabia).

However when the rock has a very poor permeability even the long wells will only very slowly accumulate fluid from the surrounding rock, since there are no easy passages to the well through that rock. Thus the next benefit that was needed was the ability to crack the rock around the well. This is known as hydraulic fracturing or hydrofracking for short. In modern wells, by isolating and then pressurizing different segments of the well in turn, these cracks can be created (when the pressure inside the well exceeds the rock strength) at regular intervals (say 30 to 120 ft apart) along the length of the borehole.

The cracks are controlled in length (since if they go outside the reservoir all the fluid can drain away through the other end of the cracks, not to the well). But the problem is that once the crack is made, the pressure inside the well is lowered and the equipment moved to the next segment. Without any other changes as the pressure comes off the crack it will close back up, and there will not be much gain from the effort. So to keep the crack open what the industry calls a proppant, but you or I might just call it a carefully sized sand, is mixed with the fracking fluid before it is injected into the well.

As a result, when the cracks open in the rock, and the fracking fluid flows into the crack, the sand is carried with it, and is then trapped in the crack, holding it open after the pressure is lowered. A passage then exists for the gas or oil to travel to the well and production of most of the rock volume becomes possible.

Well that was when the development of the gas shale deposits began, however it had not been going on very long when it was noticed that the sand was not flowing easily into the fractures, and without enough sand being carried far enough back into the cracks, production wasn’t nearly as good as it should have been.

At this point another development was needed. This came about when an additional chemical – what is known as a long-chain polymer (typically a polyacrylamide) - was added to the fracking fluid. These fluids are known as Friction Reducing Agents (FRAs) because they tend to make water stick together a bit, and create extremely slippery surfaces when they coat them. By adding these FRAs to the fracking fluid, the crack walls became slipperier and the sand particles could thus travel deeper into the cracks, holding them open more effectively and increasing gas production. The fluids were given the generic name “slick water”, so that the current state-of-the-art is a horizontal well that has had a multi-fracture, slickwater-hydrofracking operation run on it.

But the problems of the wells are not over. As I noted in my post yesterday, the mudstones contain a significant amount of clays. And the problem when clays get wet is that they get softer and clay particles can break away from the wall of the fracture (slaking). Over the different gas shale deposits the problems are not consistent, since each shale is made of a different set of constituent rock types and clays. But overall the problem that is now being evidenced, as Art Berman has commented a number of times, is that the wells are losing production faster and earlier than predicted, so that they cannot meet the overall targets that make the well profitable. Instead of the well lasting perhaps a decade, they are losing perhaps 60% of the flow in the first year, and are no longer worth operating after maybe three years.

With all that as background, here is a hypothesis to explain Halliburton’s actions. It is quite possible that the well failures are due to the clay failure in the shale reducing the crack effectiveness. Clay content failure can do this, once the fracking fluid has cracked and wetted it, by a long term softening (which will allow the walls of the crack to fold around the proppant particles, and close the crack as the walls move in), or simply swelling into some of the crack space, with the same effect. Alternately the clay particles may slake and break away from the walls of the crack, and over time build up small dams along the crack path, again blocking the fluid flow through the crack – any one of these mechanisms explains the production falls that are being seen in the industry.

So lets say that Halliburton has realized the problem and, for merely the sake of a discussable solution, changes the polymer that they use from a pure polyacrylamide (PA) to include polyethylene oxide (PO). One thing that PO does at much lower concentrations than PA is that it stops the fracking fluid from wetting the shale, and interacting with the clay. Because it is (or at least was when we did this) much more expensive than PA there is not normally any reason to use PO in the fracking fluid.

But let us say that Halliburton have tried this, and it works. Because it is a step change in the process (in the same way as horizontal drilling; fracking; and slick water use were each, in turn) then the company selling the new idea has a tremendous commercial advantage. They can promise you that your well will stay in production long enough for you to make a profit, while the competition cannot.

The world of hydrofracking contractors is small and engineers move around, so that commercial advantage does not last very long, and word gets out as to how it was done. But that takes time, first to find out what is causing the problem, then what the answer is in general, and then what the answer is in detail. Each of those steps might take a competitor a year. That gives you three years of advantage, when you can charge higher rates, and possibly put some of that competition out of business.

The problem is that if the competition sees that you have put PO in your fluid, instead of PA then they can immediately go and look up what difference that makes to the fluid. Knowing that it stops wetting immediately gets them past stages one and two and cuts the term of your commercial advantage from three years to one.

Would you want to give that up if, by lawyering and all those fancy tricks they get up to in Washington you could get the time that you have to release the content postponed by at least a year? Likely not, and since dragging out the process can extend the period of your commercial advantage, the longer you can keep kicking the ball down the street the greater your advantage, and the more benefit.

And I re-iterate this is purely a hypothesis that I came up with, and I have no connections that would suggest that this has any connection to reality.