- Created: Tuesday, 15 November 2011 10:47
- Published: Tuesday, 15 November 2011 10:47
By John Getty
Everywhere you turn these days it seems there is news about fracture stimulation. Recent headlines trumpet the concerns of environmental groups about ground water contamination. It’s easy to understand their worries, when some service companies that perform fracture stimulation operations have refused to identify the chemicals they are pumping into the ground. More positive headlines announce a substantial growth in U.S. oil and gas reserves – the amount of oil and gas that can actually be produced from a formation – due almost entirely to new extraction technologies based on fracturing the reservoir rock. And even though much of the United States remains despondent over the economy, these technologies are enabling states like North Dakota to boast some of the lowest unemployment rates in the country.
It is pretty clear why there is so much buzz. No matter which way you look at the industry, the numbers associated with fracture stimulation are huge. Reports published by RigData indicate that a quarter of a million oil and gas wells were started last year, up from about 160,000 in 2009. Halliburton, one of the major players in oil and gas well stimulation, estimates that more than 80 percent of new wells are fracture stimulated. Since it is common for a well to require 100,000 lbs. or more of proppant, a conservative estimate is that billions of pounds of frac sand was pumped into the ground in 2010. . . just in the United States.
With this kind of demand nearly every part of the system is feeling the strain. Fracturing equipment and personnel are being added at record rates. Some of the materials required – such as suitable frac sand (the more technically correct terminology is fracture proppant) – are becoming so hard to find that material is being imported from places as far away as China. This kind of demand, plus significant transportation costs have been causing the price of frac sand to mushroom.
Face it, Frac it
In higher permeability formations gas and oil are produced through a perforation – a hole in the casing and cement that permits fluid from the producing zone to enter the well – which is normally created with a shaped explosive charge. These types of perforations usually measure about an inch in diameter and a foot in length. But over the last 100 years we have produced and burned much of the recoverable oil and gas from these high-permeability formations. The newest horizons, like the Bakken in North Dakota and the Marcellus shale in New York and Pennsylvania are termed unconventional because the permeability is so low that normal perforations are insufficient to produce economic quantities of hydrocarbons. It’s in these low-perm formations that fracture stimulation completely changes the economics.
Fracture stimulation is, in theory, a relatively straightforward process. Pumping frac fluid under very high pressure into a well will create a fracture system within the reservoir. With as little as 10,000 gal., and assuming the fracture is a uniform 1/4 in. in width, a surface area exceeding that of several football fields can be obtained. This large contact surface then provides a path sufficient to produce economic qualities of oil and gas even in formations with very low permeability. But for this fracture to continue to produce fluids, it must somehow be held open. To accomplish this, proppant is transported into the fracture during the pumping phase of the frac job. When the pressure is subsequently reduced and the well is brought into production, the high conductivity of the fracture is maintained by keeping it “propped” open with the sand.
In practice, however, the fracture fluid system – including any proppant – must be carefully designed. The pressure required to induce fractures in the target formation, the compatibility of the frac fluid with the formation, the mechanical limits of the well and surface equipment and the size and type of proppant all interact to determine the success of the procedure.
Is This Proppant?
With the current shortage of proppant there is a push to find and produce new sources. It turns out that certain characteristics of the material is critical to the success of the frac job. While some of these characteristics might be unfamiliar, it is useful to be able to recognize the more important ones. And, it is handy to understand the terminology used in the fracture stimulation industry.
Several different materials are used as a fracture proppant and many more have been tried and rejected. Companies such as CARBO Ceramics and Saint-Gobain manufacture a ceramic bead-like material specifically for this purpose. This man-made proppant is ideally suited since the particles are round and strong. Their uniformity and resistance to chemical decomposition lead to high long-term fracture conductivity. But they are also expensive.
Other proppants, such as plastic and cellulose products, have seen limited use because of their tendency to deform under geologic pressures. Sometimes O&G operators will specify resin-coated sand, since the resin provides some benefit with respect to the amount of sand that is produced back to the surface. The resin coating also tends to reduce the release of fines that might plug the fracture.
But the most common proppant by far is native sand. Compared to the man-made materials it is relatively inexpensive. In a properly designed fracture stimulation program, the long-term performance of good quality material is excellent. And, production of a saleable product often requires little more than sorting for size and removal of fines.
So what is it that makes a native sand suitable for this purpose? To provide reliable, repeatable methods of quantifying the characteristics of the proppant, the American Petroleum Institute (API) and the International Standards Organization (ISO) have jointly published API Recommended Practice 19C and ISO 13503-2, both named “Measurement of Properties of Proppants Used in Hydraulic Fracturing and Gravel-packing.” Commercial testing labs, such as StimLab and Proptesters, have built a booming business around evaluating samples according to these standards. These lab tests can be critical in successfully marketing the final product.
Frac Sand and the API
The highest quality native frac sand is generally characterized by single-crystal quartzite particles that, over geologic time, have experienced significant transport and therefore are well rounded and spherical. Materials containing feldspar or calcite have characteristics that make it less desirable. Feldspar inclusions tend to weaken the sand particles and typically are soluble under the rather chemically aggressive environment often experienced in oil and gas wells. Calcite, typically present as a cementitious material in sandstone, leads to “clustered” sand particles and can behave as a contaminant in the well. The proto-typical native proppant is probably the highly desirable Ottowa sand, shown in Figure 1.
API’s Recommended Practice 19C, along with the associated ISO standard 13503-2, together lay out a set of criteria that can be used in qualifying material for use as a proppant. The Proppant Research Division (PRD) of the Petroleum Engineering Department at Montana Tech uses these guidelines plus some additional tests to advise potential producers on the desirability of their product. These tests, described below, offer a method to quickly discern if a material has potential as a fracture proppant. The material must past the first several tests to be a viable candidate. Additional testing after that point provides a way to grade the desirability of the material and provide data useful in marketing the product.
The testing protocol at the PRD follows this sequence:
1. Is it quartzite? As described above, only quartzite sand has the strength and resistance to chemical decomposi- tion required for use as a proppant. Simple tests such as dropping dilute hydrochloric acid on the material helps rule out some imposters. A petrographic microscope can consistently identify quartz sand.
2. Are the grains round and spherical? This sounds like the same thing and it is true that a spherical particle is also round. However, it is possible for sand grains to have a shape like a potato, producing a high roundness factor but a lower sphericity. Section 7 of API’s RP19C specifies that a microscope image be used to select 20 particles for analysis. These are then individually graded according to the Krumbien/Sloss chart (Figure 2) and an average is taken separately for the sphericity and roundness. While API RP19C does not explicitly lay out a “go, no-go” number, older API documents set 0.6
as a defacto threshold. Higher ratings for sphericity and roundness add value to the product.
In the field, a hand-held magnifier is often sufficient to estimate these values. If under slight magnification most of the grains are pointed and broken, it is unlikely the material will have commercial value as a proppant.
3. If the material has passed the first two tests — mineralogy and sphericity/roundness — the next step is to determine the particle size distribution. Understanding the size distribution in the raw material allows an estimate of the value of the sand deposit. Typically, larger material is more valuable since the pore throats in the proppant pack are larger and therefore the fracture conductivity is higher.
API RP19C defines a convention for measuring the particle sizes of proppant material using standard mesh sieves. The size of a proppant, labeled for example 20/40 or 30/50, is determined using stacks of seven sieves so that the distribution of the particle sizes of the material can be determined. The two numbers associated with the designated proppant size refers to the first and second primaries as specified by API. For example, the sieve stack for 20/40 proppant consists of mesh sizes 16, 20, 25, 30, 35, 40, 50 plus the pan. The bold numbers in this list identify the first and second primaries. An example distribution plot for a commercial grade 16/30 is shown in Figure 3.
In an ideal proppant, every particle would be precisely the same size. Materials with a narrower size distribution seem to suffer fewer incidents of “screen out,” a condition that can occur during the frac job in which slurry flow into the formation becomes blocked. Ultimately, materials with a uniform particle size offer a higher fracture conductivity, providing a direct link between the particle size distribution plot and the eventual economics of the well.
In some cases the characteristics of the material, such as roundness and the number of clusters (grains that are cemented to each other), change with the particle size. While it is possible the product contains economic fractions within several of the API sizes, often the char acteristics of one size makes it a better, more economic candidate than others. If there is a significant fraction of the bulk material that falls into the larger size sieves, chances increase that the material will find application as a proppant. But there are some additional questions that potential buyers will be asking.
4. Percent-of-crush testing addresses the ability of the proppant to withstand closure stress in the fracture. This is important since one of the primary mechanisms that reduce the conductivity of the fracture is thought to be plugging of pore throats from migrating fines. If the original material is clean — starting out with a small percentage of fines — and it is strong enough to with- stand the closure stress without significant crushing, long-term performance is improved. All else being equal, higher strength proppant is required in deeper wells.
If your material has made it this far, it is probably worthwhile to invest in having a sample evaluated at a proppant-testing laboratory. The API recommended practices for crush testing require the use of a force- controlled load frame and a cylindrical crush cell. The ISO 13503-2 standards further extend the procedure to test the sample at loads increasing by increments of 1,000 psi until 10 percent of the sample falls through the smallest sieve into the pan.
Several factors play into the strength of the proppant. Larger material, which has fewer points of contact to transmit the load, tends to have a lower crush strength. Material that is more spherical tends to be stronger, probably because of reduced point-loading. Inclusions, dislocations and fractures in the intergranular crystals will also tend to lower the crush resistance. For all of these reasons, an optical evaluation typically provides only a hint of the actual strength of the material.
A good quality 20/40 native proppant will generally withstand a closure stress above 5,000 psi. Ceramic proppants typically boast crush resistance at stresses above 10,000 psi.
There are a number of additional tests called for by the API recommended practices and ISO standards. These include things like acid solubility, bulk density, loss on ignition and turbidity. These RP19C tests are more meaningful for material that is ready to go to market, or has special characteristics such as resin coating. With the possible exception of acid solubility, these tests usually have little impact on the final pricing of the product.
Most of these initial tests can be conducted with a relatively inexpensive suite of tools. Before engaging the services of a testing lab, it is worthwhile to do a bit of fishing around to see how your material measures up. With any luck, you might just hook a big one.