Last week wrapped another outstanding SPE Hydraulic Fracturing Technology Conference. Every year, I write a blog post highlighting notable HFTC papers. Here is this year’s edition!

As in past years, this is not an attempt to pick the ‘best’ papers. It’s a selection of papers that I personally found insightful, based on my own interests and specialization.

 

Evaluation of Completion Designs and Fracture Heterogeneity via an Instrumented Slant Monitor Well

Benish et al. (2024)

This paper describes an exceptional pressure observation well project performed by ExxonMobil in the Midland Basin. This project is the cousin to recent pressure observation well studies in the Bakken by XOM and Hess (Liang et al., 2022; Cipolla et al., 2022) and by ConocoPhillips in the Eagle Ford (Raterman et al., 2019). In these projects, the ‘observation’ wells are cased and cemented but never fractured or produced; they are instrumented with pressure gauges and other diagnostics.

Benish et al. (2024) drilled the offset pressure observation well at an angle, so that they would have pressure measurements as a function of distance from the neighboring production well. They also installed fiber in the offset well. The fiber allowed them to count: (a) the number of frac hits during stimulation, and (b) the number of depleting fractures during production.

They compared three different completion designs with alternating A/B/C stages along the lateral. The advantage of this design is that it normalizes for geologic variability. Alternatively, if the designs had been tested with “A” in the heel, “B” in the middle, and “C” in the toe, then the results could have been affected by variability in rock properties or depletion along the lateral.

Figure 9 from Benish et al. (2024) is one of the more remarkable figures that you’ll ever see in an SPE paper. As a function of distance, it shows the percentage of the wetted fractures (as seen in the fiber during propagation) that are conductive/producing (as seen in the fiber during production and subsequent interference tests).

According to the fiber, all three designs had similar ‘wetted’ length (with most fractures reaching out from 650-800 ft laterally), but there was a significant difference in the length and distribution of the conductive fractures. Their Design C performed best, with nearly 100% of the fractures conductive out to 450 ft. Their Designs A and B had a much smaller percentage of fractures that were conductive beyond a few hundred feet.

Their Figure 11 shows pressure versus time for the three designs at different distances. The results are qualitatively consistent with the results from the fiber – showing the most pressure decline with Design C and the least with Design A.

These observations are really outstanding – Benish et al. (2024) have built an exceptionally clear picture of what is happening in these wells – where are the fractures, where is depletion, and what is the distribution of the far-field depletion.

 

Tying Stage Architecture to Wolfcamp Performance

Barhaug et al. (2024)

This paper from Ovintiv describes a field pilot evaluating stage and perforation designs in the Wolfcamp. They trialed three different designs – 300 ft stage length with 0˚  perforation phasing (top shots), 150 ft stage length with 0˚  phasing (and halved cluster spacing), and 300 ft stage length with 90/270˚  phasing. The designs were alternated A/B/C down the length of the well, with each 150 ft design repeated consecutively so that it would occupy the same lateral length as the other designs. They compared two methods of estimating production by stage (oil tracer and dip-in fiber) and assessed which designs had the best production. They also ran downhole imaging with EV to observe the distribution of perforation diameters before and after pumping.

The downhole imaging (EV) observations provide measurements of perforation erosion. They observe a large difference in initial hole diameter between the 0˚  and 90/270˚  designs (.28’’ versus .37’’), which implies a large difference in the limited-entry perforation pressure drop (a roughly 3x difference). Interestingly, they report that the perforation erosion was ‘uniform’ for the two cases, which would evidently imply a relatively uniform outflow of fluid and proppant from the wellbore in each cluster.

In the production comparison, they found that the 90/270˚  design had the lowest production. Comparing the top-shot designs, there was some discrepancy. The oil soluble tracer diagnostics found that production was greatest from the 300 ft top shot stages. Conversely, the dip-in fiber found that the greatest production was from the 150 ft stages.

This is a thought-provoking set of results. Some questions:

1. If the perforation erosion was ‘uniform’ between the designs, why were there differences in production? The 300 ft stage top shot and 90/270˚ designs were identical, except for phasing. The difference in phasing affects: (a) initial shot diameter and the magnitude of limited-entry, and (b) proppant outflow from the well. Both of these factors should affect the uniformity of flow from the well. But if the diagnostics suggest that the phasing did not affect the uniformity of erosion (and outflow) from the well, then there isn’t an obvious mechanism to explain why they would have led to different production. A few possible explanations: (a) near-wellbore tortuosity significantly affects production and systematically varies with perforation phasing, or (b) the results are affected by randomness or imprecision in the production allocation diagnostics. To further evaluate, it would be interesting to do a more detailed analysis of the erosion data, such as in Dontsov et al. (2024), discussed below.
2. As noted above, the top-shot design had much smaller initial shot diameter than the 90/270˚ This difference in initial diameter could have an effect on the distribution of flow from the well. It would be interesting to repeat these diagnostics while controlling to use the same initial diameter at all shot orientations.
3. It would be interesting to trial pure 90˚ phasing, rather than alternating sides with 90/270˚ . It may be advantageous to have consecutive shots on the same side, because if proppant ‘misses’ the first shot, it may be well-positioned to outflow from the subsequent shots.

Overall, a fascinating set of results. I really like their alternating A/B/C design and use of overlapping diagnostics to compare the results.

 

Statistics and Case Studies of Drainage Frac Height and Zonal Contribution of Stacked Plays in the Midland and Uinta Basin

Ge et al. (2024)

This paper provides a statistical summary of RevoChem geochemical analyses of drainage patterns in the Midland Basin (269 wells) and Uinta Basin (118 wells). The basis of RevoChem is that the chemical composition of the hydrocarbon mixture in each layer is slightly different. By analyzing the overall composition of the fluid produced at the surface, it is possible to calculate the percentage of production coming from each layer.

Before we dig into the results, we should note that stratigraphy, stress profiles, and rock quality are nonuniform across a basin, and this leads to significant variability in height growth and drainage. We cannot look at basin-wide summary statistics and assume that they apply directly to specific projects. Nevertheless, statistical summaries like this paper are very valuable.

Ge et al. (2024) estimate an average producing height of 290 ft in Midland Basin wells, with P5/P95 range of 100 to 400 ft. They report considerable differences in productive height by layer, with the greatest productive height from Jo Mill, Lower Sprayberry, and Wolfcamp A, and the least productive height from the Middle Sprayberry and Wolfcamp B, C, and D.

In the Uinta Basin, they estimate an average of 208 ft, with P5/P95 range of 100 to 300 ft. They observe relatively greater productive height from wells landed in the Wasatch formations, with lesser productive height from wells in the Uteland Butte and Castle Peak landing depths.

Ge et al. (2024) also provide statistics on average formation thickness in the two basins. The formations are generally thicker in the Midland Basin (220-400 ft, except for the especially thick Wolfcamp C at 559 ft) than in the Uinta Basin (140-280 ft).

As I reviewed these results, I felt that they were concordant with my own expectations based on past experience. Midland Basin wells experience a large amount of height growth due to the relative paucity of high-stress frac barriers (McClure et al., 2023).

 

Cementing: The Good, the Bad, and the Isolated – Techniques to Measure Cement Quality and its Impact on Well Performance

Haustveit et al. (2024)

This paper provides a deep dive on the critical – but often overlooked – role of cementing on plug and perf fracturing. Plugs and perforation pressure drop are used to force fluid and proppant to exit the wellbore relatively uniformly from the well. Uniform flow out of the casing is supposed to create a uniform placement of fluid and proppant in the far-field away from the well. However, this strategy can be defeated if there is a large amount of cross-flow outside the casing. Cross-flow outside casing could happen because of the formation of a longitudinal crack or because of insufficient cement quality. A certain amount of cross-flow outside casing is likely unavoidable. However, if there is severe cross-flow, it could have a significantly detrimental effect on the uniformity in the far-field.

Haustveit et al. (2024) measure annular isolation with a bottomhole assembly that places pressure gauges on either side of the bridge plug. This allows them to observe whether the prior stage experiences anomalous pressurization during fracturing of the subsequent stage. They compare the results with ultrasonic imaging tools (USITs) that image the cement annulus and – reassuringly – find that the USIT measurements are effective at predicting the cement quality and degree of ‘outside casing’ isolation from the prior stage. They further validate their results with RA tagged proppant and DAS.

What are the keys for achieving good cement placement? Haustveit et al. (2024) find that the number of centralizers per casing joint and the cement annular velocity during pumping are the two most important for cement quality. On the topic of rotating while cementing, they say:

“While casing rotation during cementing improves the radial coverage of the cement, the cost of high torque threads can exceed the expected value of the operation. In addition, longer laterals with more complex wellbore geometries increase the level of difficulty associated with rotating casing during cementing.”

Finally, Haustveit et al. (2024) relate the assessments of cement quality to production. As expected, they find that wells with better cement quality perform better. However, this result appears to depend on the diameter of the annulus. They say: “Small annulus wellbore constructions are less reliant on a high-quality cement bond because the pressure drop is large if a slurry begins to move up or down the annuls.”

Another interesting paper on zonal isolation was presented by Savitski et al. (2024). They compared stage isolation estimates between ‘pump down diagnostics’ (aka, isolation integrity tests) and fiber optic. They found that the pump down diagnostic assessments were overly conservative. In some cases, the tests indicated communication with the prior stage, but that result was not confirmed by the fiber measurements.

 

A Comprehensive Review of Casing Deformation During Multi-Stage Hydraulic Fracturing in Unconventional Plays: Characterization, Diagnosis, Controlling Factors, Mitigation and Recovery Strategies

Uribe-Patino et al. (2024)

For anyone interested in understanding casing deformation in unconventionals, this paper is a fantastic starting point. This review paper was written by the Casing Deformation Work Group in the SPE Well Integrity Technical Section. They categorize four root causes of casing deformation – (a) issues with casing and tools, (b) cement quality, (c) localized tectonics (such as fault slip), and (d) operational issues. Casing deformation causes ‘wellbore access’ issues that make it difficult to place plugs and mill them out. Uribe-Patino et al. (2024) documents these challenges, along with strategies for prevention and mitigation.

 

Practical Optimization of Perforation Design with a General Correlation for Proppant and Slurry Transport from the Wellbore

Dontsov et al. (2024)

This paper from ResFrac reviews case studies using our StageOpt tool (resapps.resfrac.com). The tool is a fast-running wellbore dynamics simulator for predicting fluid and proppant outflow from the well and perforation erosion. The simulator implements physics-based relations for fluid and proppant transport, which were developed by Egor Dontsov, drawing on the existing literature. The purpose is to optimize perforation design (spacing, shot count, and phasing) to maximize uniformity index and maximize production.

Dontsov et al. (2024) performed generic sensitivity analysis simulations and applied StageOpt to two field case studies. The sensitivity analysis simulations clarify the relative effect of different processes. Turning on or off different effects, the paper maps out how stress shadow within the stage, stress shadow from the prior stage, erosion, inertial, and gravitational effects impact the tendency for heel or toe bias.

In the two field case studies, we show how the model can be applied to match field data. After matching, the model is used to optimize the perforation phasing and the number of shots per cluster. We can optimize: (a) phasing, (b) shots per cluster, (c) initial shot diameter, and (d) whether or not to use tapered perforating.

I am really excited about this new tool. We’ve captured the key processes driving proppant transport from the well, and packaged them into an easy-to-use, fast-running tool for modeling field data and evaluating alternative scenarios.

 

Other notable papers

Dreyer et al. (2024) and Sun et al. (2024) describe the development of frac fluid formulations designed to minimize conductivity damage from chemical interactions. As discussed by Ratcliff et al. (2022) and McClure et al. (2023), frac fluid damage is an especially big deal in the Montney and Scoop/Stack. But even in other formations, it can have a significant impact on production.

Titov et al. (2024) provide an update on the really exciting work that Fervo Energy has been doing applying multistage hydraulic fracturing for geothermal energy. One notable aspect of this paper – they used the interference testing method from Almasoodi et al. (2023) to estimate the fracture conductivity between the wells and estimated between 68-164 md-ft for each fracture. In-situ propped fracture conductivity has long been viewed as one of the key uncertainties when projecting the potential power production from multistage EGS designs. The field results from Titov et al. (2024) provide conductivity estimates that are very favorable.

For comparison, in McClure et al. (2022), we assumed a baseline conductivity of 40 md-ft, with an upper limit ‘optimistic’ value of 200 md-ft. In Li et al. (2016), we assumed a baseline value of around 1000 md-ft per stage; assuming eight fractures per stage, this would be equivalent to 125 md-ft for each fracture.

In another ResFrac paper, Morsy et al. (2024) look at the impact of well orientation on well productivity in the Bakken. Intuitively, we expect that drilling in the direction of Shmin maximizes productivity. However, because of the orientation of leases, this is not always possible. How much does this matter?

Morsy et al. (2024) start by reviewing a statistical analysis from Roostami et al. (2020). The results are generally intuitive – drilling on-azimuth does best. However, the statistical review is hampered by confounding covariates. Differences in underlying parameters within the dataset (such as job size and the presence of offset parent wells) make it difficult to make a clean comparison. Morsy et al. (2024) performed sensitivity analysis simulations to disentangle these processes.

The paper says that “simulation sensitivity analyses show that depletion, wider cluster spacing, and wider well spacing lessen the effect of well orientation on well productivity.” These results help explain why some practitioners insist that orientation makes a big difference, and others insist that it does not. As is often the case, these perspectives might both be right – it depends on the circumstances. Modeling can help explain when and why.

 

References

Almasoodi, M., Andrews, T., Johnston, C., Singh, A. and McClure, M. 2023. A new method for interpreting well-to-well interference tests and quantifying the magnitude of production impact: Theory and applications in a multi-basin case study. Geomech. Geophys. Geo-energ. Geo-resour. 9 (95).

Benish, T. G., J. S Brown, S. S. Chhatre, K. K. Decker, J. D. Habrial, R. Manchanda, A. Solomou, and D. S. Vice. 2024. Evaluation of Completion Designs and Fracture Heterogeneity via an Instrumented Slant Monitor Well. SPE-217825-MS. Paper presented at the SPE Hydraulic Fracturing Technology Conference, The Woodlands, TX.

Cipolla, C., J. Wolters, M. McKimmy, C. Miranda, S. Hari-Roy, A. Kechemir, and N. Gupta. 2022. Observation Lateral Project: Direct Measurement of Far-Field Drainage. SPE Hydraulic Fracturing Technology Conference and Exhibition, The Woodlands, TX.

Dontsov, Egor, Christopher Ponners, Kevin Torbert, and Mark McClure. 2024. Practical Optimization of Perforation Design with a General Correlation for Proppant and Slurry Transport from the Wellbore. SPE-217771-MS. Paper presented at the SPE Hydraulic Fracturing Technology Conference, The Woodlands, TX.

Dreyer, Daniel, David Garza, and Pious Kurian. 2024. Evaluation of the Performance of Chemical Breakers on Various Types of High Viscosity Friction Reducers. SPE-217788-MS. Paper presented at the SPE Hydraulic Fracturing Technology Conference, The Woodlands, TX.

Ge, Shuangyu, Jana Bachleda, Luke Fidler, Josh Sigler, Tao Lv, and Faye Liu. 2024. Statistics and Case Studies of Drainage Frac Height and Zonal Contribution of Stacked Plays in the Midland and Uinta Basin. SPE-217764-MS. Paper presented at the SPE Hydraulic Fracturing Technology Conference, The Woodlands, TX.

Haustveit, Kyle, Jackson Haffener, Steven Young, John Dwyer, Garrett Glaze, Brett Green, Chris Ketter, Tyler Williams, Kourtney Brinkley, and Brendan Elliott. Cementing: The Good, the Bad, and the Isolated – Techniques to Measure Cement Quality and its Impact on Well Performance. SPE-217797-MS. Paper presented at the SPE Hydraulic Fracturing Technology Conference, The Woodlands, TX.

Li, Tianyu, Sogo Shiozawa, and Mark W. McClure. 2016. Thermal breakthrough calculations to optimize design of a multiple-stage Enhanced Geothermal System. Geothermics 64, 455-465.

Liang, Y., and Meier, H., Srinivasan, K., Tahir, H., Ortega, A., Amalokwu, K., Friedrich, J., Okpala, J., Decker, K., Brown, M., Benish, T., Crozier, H., Bhargava, P., Sumant, P. and Reavis, J. 2022. Accelerating Development Optimization in the Bakken Using an Integrated Fracture Diagnostic Pilot. URTEC-2022-3719696-MS. Paper presented at Unconventional Resources Technology Conference, Houston, Texas.

McClure, Mark, Charles Kang, and Garrett Fowler. 2022. Optimization and Design of Next-Generation Geothermal Systems Created by Multistage Hydraulic Fracturing. SPE-209186-MS. Paper presented at the SPE Hydraulic Fracturing Technology Conference and Exhibition, The Woodlands, TX.

McClure, Mark, M.L. Albrecht, C. Bernet, C.L. Cipolla, K. Etcheverry, G. Fowler, A. Fuhr, A. Gherabati, M. Johnston, P. Kaufman, M. Mackay, M.P. McKimmy, C. Miranda, C. Molina, C.G. Ponners, D.R. Ratcliff, J. Rondon, A. Singh, R. Sinha, A. Sung, J. Xu, J. Yeo, R.B. Zinselmeyer. 2023. Results from a Collaborative Industry Study on Parent/Child Interactions. SPE-212321-MS. Paper presented at the SPE Hydraulic Fracturing Technology Conference and Exhibition, The Woodlands, TX.

Morsy, S., G. Fowler, M. McClure, and D. Ratcliff. 2024. Impact of Well Orientation on Well Productivity. SPE-217782-MS. Paper presented at the SPE Hydraulic Fracturing Technology Conference, The Woodlands, TX.

Raterman, Kevin T., Yongshe Liu, and Logan Warren. 2019. Analysis of a drained rock volume: An Eagle Ford example. URTeC-2019-263. Paper presented at the Unconventional Resources Technology Conference, Denver, CO.

Rostami, Erfan, Boness, Naomi, and Mark D. Zoback. 2020. Significance of Well Orientation on Cumulative Production from Wells in the Bakken Region. Paper presented at the SPE/AAPG/SEG Unconventional Resources Technology Conference, Virtual

Savitski, Alexei A., Benjamin Lock, Felix Todea, Patricio Fita, Gabriel Formento, Hawraa I. Al Lawati, Somnath Mondal, and Gustavo Ugueto. 2024. Improving Stage Isolation in Vaca Muerta Wells Through Observations from an Integrated Fiber Optic Pilot. SPE-217785-MS. Paper presented at the SPE Hydraulic Fracturing Technology Conference, The Woodlands, TX.

Sun, Hong, Ying-ying Lin, Xi Geng, Lanka Wickramasinghe, Fulya Zalluhoglu, and Qing Wang. 2024. Engineering a Synthetic Friction Reducer to Combat Undesirable Formation of FR-Metal Complex/Precipitation in Slickwater Fracturing. SPE-217763-MS. Paper presented at the SPE Hydraulic Fracturing Technology Conference, The Woodlands, TX.

Titov, A., S. Dadi, G. Galban, and J. Norbeck, M. Almasoodi, K. Pelton, C. Bowie, J. Haffener, and K. Haustveit. 2024. Optimization of Enhanced Geothermal System Operations Using DistributedFiber Optic Sensing and Offset Pressure Monitoring. SPE-217810-MS. Paper presented at the SPE Hydraulic Fracturing Technology Conference, The Woodlands, TX.

Uribe-Patino, J. A., A. Casero, D. Dall’Acqua, E. Davis, G. E. King, H. Singh, M. Rylance, R. Chalaturnyk, and G. Zambrano-Narvaez. A Comprehensive Review of Casing Deformation During Multi-Stage Hydraulic Fracturing in Unconventional Plays: Characterization, Diagnosis, Controlling Factors, Mitigation and Recovery Strategies. SPE-217822-MS. Paper presented at the SPE Hydraulic Fracturing Technology Conference, The Woodlands, TX.