Proppant power.

Larger-diameter proppant will hold the crack surfaces farther apart, and the width of the crack will be larger and allow a more rapid flow of gas or oil down the conduit to the well (Figure 1). But there’s a two-fold limit: the proppant-sand has to be light enough to be carried by the frac water flow and it also has to fit into the created cracks.

20-40 proppant (diameter about 0.5 mm) evolved as the optimum proppant in most gas and oil wells, until the resource known as coalbed methane entered the scene in the 1980s. Coal contains large amounts of natural gas but to get it out almost all wells had to be fracked.

In the San Juan basin, operators were using gelled water (thickened water) to better carry the standard 20-40 proppant-sand, but in some locations, they couldn’t get the desired amount of proppant into the well because it blew up the pressure and they were afraid the well would burst – the sand grains were too big to fit into the cracks.

The research lab I worked in suggested starting out with a smaller 40-70 sand (0.25 mm diameter) then graduating to the standard 20-40 sand. Problem solved!

Meanwhile in Warrior basin coals, one sharp engineer in the early 1990s suggested not using any proppant at all and using plain water as the frac fluid instead of gelled water. The fracking operation was cheaper and initial gas flow rates were pleasingly high. But these flow rates eventually dropped to such a low level it became clear that proppant-sand was needed in coalbed methane wells, as a lot of wise fracking engineers already knew.

The stage was set for shale wells.

After coalbed methane, the next unconventional resource was shale, and proppant became an issue there also. George Mitchell’s company was working the Barnett Shale, out of Fort Worth. After many failures, in the late 1990s they tried plain water as the frac fluid, and smaller-diameter sand (which they called “light sand”), and wells became commercial.

They also tried horizontal wells but it was 2003 before Devon, who bought out Mitchell, fully embraced the technology of long horizontal wells fracked many times along its length. They used plain water and a mix of 100-mesh sand (0.15 mm) ahead of 40-70 sand (0.25 mm diameter).

The secret of success was that this approach created a network of cracks (sub-fractures) surrounding the horizontal well along its total length. The gas and oil molecules that can barely move through raw shale rock, can hustle along if they can get into a tiny crack – and now there were many such cracks. One study showed that almost 50% of oil flow came from secondary cracks in the network.

With such a network of cracks, it stands to reason that the opening width of such cracks will be a lot smaller than the width of a single vertical fracture as shown in textbooks. Therefore, the standard 20-40 proppant-sand can’t be pumped, but smaller 100-mesh and 40-70 sand can fit into these cracks. It all makes sense, except that nobody knew how wide these tiny cracks were.

In 2014, a step was taken to understand this. Five wells in the Fayettville shale, in Arkansas, were studied (Ref 1) and the width and spacing of cracks in the fracture network around each of these wells were calculated. It was a complicated method – let’s just say a geomechanics model was developed to match the pattern of microseismic signals which record cracking of the shale rock caused by the water pressure.

The widths of the cracks in the network were consistent with the sizes of 100-mesh and 40-70 proppant-sand used in the shale fracking operations. This verified the model used to match the microseismic data, and for the first time explained why 20-40 proppant-sand could not be used to frac shale wells – it couldn’t fit into the cracks.

Another aspect of this work was to analyze production data from the same five shale-gas wells. Correlations were found that showed 100-mesh sand was more beneficial than 40-70 sand.

This study led to a suggestion that 200-mesh sand (0.1 mm) might give even better production than 100-mesh sand because 200-mesh sand could get into cracks that had less width than 100-mesh sand diameter. It was well known that a wide spectrum of crack widths was present in shale before and after a fracking operation.

However, one report indicated that more than 40–70 mesh was better. They rationalized that 40–70 mesh proppant-sand was better for keeping cracks in the network open because it was stronger than 100 mesh sand and more resistant to crushing by intense earth stresses trying to close the fractures.

By 2018, shale fracking operators were pushing the limits. They placed up to 40 separate frac jobs in a 2-mile long horizontal well. The total proppant-sand pumped was around 20 million lbs – enough to fill 90 railroad containers. This led to a shortage of proppant-sand, and new sand mines were opened up in Texas and Oklahoma to fill the need.

The previously unheralded role of proppant-sand quietly became a significant and expensive ingredient of fracking recipes in the shale revolution.

This is the kickoff point for a following article that looks at new data from shale wells that used even finer proppant-sand than 200-mesh – called DEEPROP. Some wells revealed a significant uplift in gas and oil flowrates.

Ref 1: Palmer et al., “Case Histories from Fayetteville Shale: SRV Sizes, Fracture Networks, Spacing, Aperture Widths, and Implications for Proppant,” SPE 169015, 2014.