Pumping proppant down a wellbore is the easy part.
Ensuring that the precious material does its job is another matter.
A trio of field studies recently presented at the 2020 SPE Annual Technical Conference and Exhibition (ATCE) highlight in different ways how emerging technology and old-fashioned problem solving are moving the industry needle on proppant and conductivity control.
These examples include the adoption of unconventional completion techniques in a conventional oil field in Russia and work to validate the use of small amounts of ceramic proppant in North Dakota’s tight-oil formations. Both studies seek to counter widely held assumptions about proppant conductivity.
A third study details a recently developed chemical coating that Permian Basin producers are applying “on the fly” to sand before it is pumped downhole. The new adhesive material has found a niche in helping operators mitigate the amount of sand that returns to surface during flowback, a sectorwide issue that drives up completion costs and later may spell trouble for artificial lift systems.
Disproving “The Overflush Paradigm”
After conventional reservoirs are hydraulically fractured, both from vertical and horizontal wells, it has been standard practice for decades to treat the newly propped perforations with a gentle touch.
The approach to this end is known as underflushing.
When underflushing, the goal is to leave behind just a few barrels’ worth of proppant-laden slurry over the perforations before attempting to complete further stages. The motivation for this boils down to the need for an insurance policy against displacing the near-wellbore proppant pack and causing the open fracture face to pinch off before it ever has a chance to transmit hydrocarbons.
Such carefulness comes at a price.
Underflushing raises the risk of needing a cleanout before oil can flow optimally to surface. This not only delays the arrival of first oil, it means extra equipment and personnel are required.
However, a more glaring downside to underflushing is that it appears to be an unnecessary precaution. The near-wellbore fracture area is, in fact, more robust than what conventional wisdom allows credit for.
This is the central argument raised by Shell and its joint-venture partnership with Gazprom Neft—the Salym Petroleum Development—in their latest technical paper (SPE 201666) detailing a multiwell study from the Salym oil fields located in southern Siberia.
“This belief about fracture connectivity has led to what we call the underflush paradigm,” explained Sanjay Vitthal, the paper’s lead author and a principal technical expert in hydraulic fracturing with Shell.
The senior engineer emphasized during his paper presentation at ATCE that the joint-venture’s experience has shown that at least in their field, this paradigm is clearly “flawed” and based on several “unrealistic physical assumptions.”
The industry’s long-held concern is “that any fracture overflush will result in lost fracture-to-wellbore connection and poor well productivity—therefore, fractures must be underflushed,” he added, while noting, “For such a strongly held belief, there is surprisingly little data in the literature to support the underflush paradigm.”
The confident assertion is based on what the two operators believe is the largest underflush vs. overflush study of its kind in a conventional setting, which in this case involves a sandstone reservoir with a permeability of up to 20 millidarcy.
In support of their argument, images from downhole cameras show in one overflushed well that fractures were wide open even though no proppant was visibly detected.
It is believed that the strong force of proppant erosion during the stimulation etched enough rock material from the fracture face to prevent it from sealing back up after fracture closure.
The revelation has led Shell and Salym Petroleum to adopt overflushing as a standard practice in their jointly operated field. Using what they call a “managed overflush approach” on more than 20 wells to date, they reported an unspecified reduction in overall well cost, lower operational complexity, and no more cleanouts.
Their breakthrough came after operational problems led a well with tracers pumped downhole to be partially completed with overflushing.
Three of that well’s seven stages were overflushed by large volumes, ranging from more than 38 bbl to nearly 150 bbl. The stage-level tracers that returned to surface during flowback indicated that despite the overflushing, none of the stages had lost their fracture-to-wellbore connection.
Later, the two operators decided to execute an intentional 20-bbl overflush to see if the results could be repeated. Through slickline and a pressure buildup test, they confirmed that while there was no more proppant sitting over the open perforations, the fractures were indeed still connected to the wellbore.
Comparison of 20 months of production data from underflushed wells on the same pad showed all wells had nearly identical flow rates and cumulative output. No evidence of productivity degradation over time was found.
The operators looked outside of this test pad and found the same pattern.
One notable exception, though, is that not one of the overflushed wells fell into the bottom quartile in terms of productivity. Furthermore, the productivity index shows that the overflushed wells dominated the top 50% of performers.
And though this is a conventional project, one of the chief arguments the petrotechnical team is using to support its findings comes from the shale revolution which has delivered hundreds of thousands of wells using the overflushing technique.
Tight-oil producers routinely pump downhole as much as 50 to 100-plus bbl of fluid to place the fracturing plugs and perforation guns needed to complete subsequent stages. Shell and Salym Petroleum point out that overflushing is not only widely considered by the shale community to be of no risk to conductivity, it has been linked to stronger well performance.
The two operators have not delivered a verdict on whether overflushing in their conventional patch of rock yields better wells. Despite the early indicators, Vitthal said more data are needed to be conclusive.
| Key Conditions Required for OverFlushing Paradigm To Be True |
|
| The operators’ analysis found that none of these factors are likely to represent true downhole or reservoir conditions. Each aspect is covered in greater detail in the full paper (SPE 201666). |
A Little Bit of Ceramics Can Go a Long Way
When it comes to proppant selection across most tight-oil developments in the US, the standard practice is relatively simple: pick one type of sand (usually 100 mesh) and then use a lot of it.
Fractures, however, are not so simple.
“A one-size-fits-all philosophy of proppant selection has permeated the entire industry,” said Mark Pearson, CEO of tight-oil producer Liberty Resources. “And quite frankly, it doesn’t really have a technical basis since the requirements for near-wellbore conductivity are very different than the requirements of the body of the fracture.”
Pearson issued his commentary during a presentation of a technical paper (SPE 201641) he coauthored with other three other firms—ResFrac, US Ceramics, and PropTester—which aided in Liberty’s recent investigation into proppant conductivity.
The portrayal of hydraulic fractures as heterogeneous systems helps sum up why Liberty prefers to supplement its sand volumes with small amounts of high-strength, high-conductivity ceramic proppants during the beginning of a fracture stage or at the end. These so-called “lead-in” and “tail-in” approaches are well established; however, Pearson noted that neither is widely used across the unconventional spectrum.
One of the reasons others have opted out is because of the perceived high cost of ceramics and skepticism over the true scope of the production benefits they deliver. Though it needs no convincing, Liberty launched the study to gain a firmer grasp on past results and to refine its execution.
The chief finding is that by adding 5 to 10% of ceramic proppant by volume as a tail-in or lead-in, the operator can expect to boost initial 180-day production output by 5 to 9% vs. an all-sand fracture design.
Under these conditions, the additional cost of the ceramics can be recovered within the first 1 to 2 months of the well’s life, according to Pearson. “These designs are increasing our free cash flow out of the well [by] half-a-million to three-quarters of a million dollars,” over a 3-year production period, he added.
The paper notes that using high-conductivity ceramic proppant to accelerate the return on investment is driven in part by a need to improve the performance of Liberty’s middling performers. The company’s best wells pay out within 18 months, while most of its inventory needs twice that time to attain breakeven status.
The economic figures shared in greater detail within the full paper are based on history matching of production data from Liberty’s assets in northern North Dakota. Liberty also used a large-scale proppant-testing system to validate its learnings along with the integrated reservoir-and-hydraulic fracture simulator first introduced by ResFrac in 2018.
Importantly, the comparisons also rely on the assumption that all-sand fracture designs suffer from some degree of conductivity damage, ranging from 50 to 90% over a 3-year time frame. Though the degree to which conductivity damage affects production remains an area of open debate in the industry, Pearson said the damage factors used here are based on published industry studies.
Based on its simulation work, Liberty has also concluded that tail-in designs are ideal in scenarios where fracture height extends above the target formation but into a nonproducing layer of rock. In the reverse, where fracture height will lead to the production of multiple zones, the operator found that lead-in designs deliver greater production.
A Sticky Solution for Proppant Flowback
It may take anywhere from 10 to 15 million pounds of sand to complete a horizontal well in the Permian Basin that spans the flatlands of Texas and New Mexico. Such volumes mean that even if a small percentage re-enters the wellbore, then many thousands of pounds of material are not where they should be.
The ensuing problems vary in complexity and cost.
Aside from concerns over the status of near-wellbore conductivity, other serious issues include reduced life spans for expensive-but-sensitive electrical submersible pumps (ESPs).
Too much sand will also cause erosional damage to wellhead chokes and flowlines. Smaller headaches, but still costly, include lengthy flowback operations and the separation and disposal of the returned sand.
Chemical specialist Hexion is among those that have explored the root cause of these issues and its answer is a proppant-flowback-control additive (PFCA) that it reports Permian operators are adopting to great effect.
Logan Cabori, a sales manager with Hexion and former field engineer, explained during his presentation of the company’s new paper (SPE 201372) that “technical shortcomings, application inflexibility, and cost of the existing techniques and technologies led to the development of PFCA.”
Introduced in 2018, the PFCA is described as a water-insoluble liquid adhesive that bonds sand particles together as they file into the tight-rock fractures. Hexion highlights that the chemical coating is added “on the fly” to the blender units (from 1 to 2% of total volume) found on all fracturing fleets where it attaches to the sand, leaving no residual coating on the equipment.
In one of the field trials, a Permian operator wanted to compare the PFCA to resin-coated proppant (RCP), an incumbant technology for this application which has been in use since at least the 1980s. The two materials were pumped down different wells in a tail-in design that represented 10% of the total volume of 100-mesh regional sand.
During the plug drillout phase, the PFCA well produced 50% less proppant than the resin-coated well. A further 80% reduction in proppant returns was seen during the flowback stage.
Cabori noted that this undisclosed operator has since adopted PFCA for all its completions and, as a result, new wells are showing an increase in cumulative production over RCP wells on the order of tens of thousands BOE.
After using PFCA in 80 wells, the paper reports, this operator saved nearly $10 million in about a year’s time as compared with the cost of its previous RCP designs.
Another Permian operator tapped Hexion for a field trial after suffering proppant-induced damage to its ESPs. This was despite attempts to stop the unwanted migration by using 25% tail-ins of RCP.
“In this case, we were able to run a 50% tail-in of PFCA while remaining cost neutral to that 25% RCP design,” said Cabori, adding, “During the first month of initial flowback, the PFCA well produced only 30 pounds of proppant to the surface.”
Further trials with this operator focused on stepping down the volume of coating to 40% and 35%. “Even at these lower loadings, only trace amounts of proppant flowback were observed,” said Cabori.
In the final case study shared, another operator was using a stimulation design comprising two different sizes of sand—100-mesh and the larger 40/70-mesh grains.
When it was realized that the majority of proppant returned to surface was 40/70 sand, the decision was made to coat this grain size as it was pumped at different intervals. The job was then finished with a 40% tail-in of PFCA-coated sand.
Initial results showed wells completed with this approach saw 75% less proppant flowback as compared to the control wells.
The PFCA has so far proven compatible with most types of fracturing fluids—including the most popular slickwater designs. Cabori said the lone exception is fluid systems using cationic friction reducers which contain acidic components that force the chemical coating to precipitate out of solution.
For Further Reading
SPE 201666 Successful Overflushing of Hydraulic Fractures Using Crosslinked Fluids in Conventional Permeability Reservoirs: Theoretical Basis for a Paradigm Shift with Field Results by Sanjay Vitthal, Shell Exploration and Production Company.
SPE 201641 Near-Wellbore Deposition of High-Conductivity Proppant To Improve Effective Fracture Conductivity and Productivity of Horizontal Well Stimulations by Mark Pearson, Liberty Resources LLC.
SPE 201372 On-the-Fly Proppant Flowback Control Additive by Logan Cabori, Hexion Inc. et al.
