Skip to main content
European Commission logo print header

Environmentally best practices and optmisation in hydraulic fracturing for shale gas/oil development

Periodic Reporting for period 2 - BESTOFRAC (Environmentally best practices and optmisation in hydraulic fracturing for shale gas/oil development)

Reporting period: 2019-01-01 to 2022-10-31

Current HF practices of unconventional hydrocarbon extraction are based on the emergence of several key technologies: the development of directional drilling, the use of high volumes of fracturing fluid, improvements in fluid lubrication as well as the use of multi-well pads. Lately, the use of HF in unconventional oil and gas extraction has generated a lot of controversy. Opponents of HF claim that its use poses severe environmental risks such as contamination of groundwater resources, that it depletes freshwater supply and induces seismic activity. There are serious environmental issues, which include possible groundwater contamination as well as risk of spills of chemicals and wastewater during vehicular transport and from on-site storage equipment. This is complicated by the fact that full disclosure of chemicals used in HF has not yet been achieved in industry despite intense public demand, so that monitoring of surface contamination is made difficult in the sense that monitoring personnel do not know exactly what substances to test for.

It is apparent that successful optimization of HF process variables requires a good understanding of the mechanics of HF treatment operations in NFRs such as shale. These variables include fluid injection pressures and rates, chemical compositions of the HF fluid, well lateral orientations and spacings, number and distance between HF stages. Ideally, we want to maximize stimulated volumes while minimizing the volume of HF fluids as well as their chemical content. Successful optimization is predicated on the ability to accurately model the physics of the HF process as well as the ability to do proper sensitivity analyses with respect to the process variables. Direct monitoring evidence suggests that fracture in an NFR evolves in a complicated manner according to the presence of local heterogeneities, layering and natural fractures. Therefore, it is also important to be able to take into account existing natural fracture networks in the numerical model.

This project aims to bring together the complementary expertise of research groups to gain a better understanding of the physics in hydraulic fracturing (HF) with the final goal to optimize HF practices and to assess the environmental risks related to HF. This requires the development and implementation of reliable computational models of HF and laboratory experiments to validate these models. The scientific objectives can be summarized as follows:
1. To generate new models to the HF analysis and control problems through exposure to different methodologies.
2. To validate the models in a stochastic sense and carry out uncertainty analysis (UA) to identify the key input parameters (such as tectonic stress, material parameters etc.) of the underlying model with respect to (w.r.t.) a pre-defined output such as the crack density, fracture volume or pressure drop.
3. To build a database and experience sharing platform for the current HF models, geological data and operating conditions.
The consortium has reached all scientific objectives and all deliverables and milestones. The consortium has developed several computational approaches for hydraulic fracturing including discrete frack approaches such as cohesive elements, extended finite element and meshfree methods and peridynamics. These approaches allow the fluid flow through discrete cracks. However, they are difficult to implement and only suitable for a moderate number of cracks. One key issue in these approaches is the lack of physical criteria to trigger complex fracture patterns such as crack coalescence and crack branching. On the other hand, continuous approaches for fracture have been developed in the consortium including phase field models for fracture and semi-analytical approaches. Those methods smear the crack over a finite width and are not capable of modeling fluid flow directly through propagating of cracks. Fluid flow is rather modelled by diffusion. Such approaches are comparatively simple to implement. Furthermore, methods for uncertainty analysis have been fully established in the consortium and at least partially applied to determine the influence of uncertain input parameters w.r.t uncertain output parameters for some models available in the consortium. Also, the datashare platform developed by our Chinese partners at Tongji University (TJU) has been set up and introduced to the entire consortium. The contents including the data and the use of this platform has been provided inside the consortium. Therefore, an additional workshop has been introduced at Tongji University in the summer as agreed by the project officer. At this workshop, the advancements and status of the project has been discussed and elaborated. Also, future events have been discussed as planned in the Annex of this project. By the end of last year, the implementation of the data-platform has been completed by our partners at TJU and the data-platform needs to be fulfilled with data. First data has already been made available by our partners in China.

Concerning exchanges and activities inside the consortium, we notice that all workshops have been carried out as intended though the secondments are delayed for several reasons including visa issues. The first workshop training the ESRs and ERs has been held successfully in Udine as intended. The mid-term meeting was held in June at Leibinz University Hannover (LUH) as requested by the project officer. The external expert evaluated the progress positively and requested minor revisions to the progress report and deliverables. These have been incorporated and resubmitted.
The progress beyond the state-of-the-art achieved so far can be summarized as follows:
1. One of the first (stochastic) 3D discrete crack models for hydraulic fracture taking into account the interaction between deformation and fluid.
2. One of the first 3D multiphysics HF models implemented in COMSOL.
3. One of the first platform for sharing knowledge, expertise and data of HF activities.
4. One of the first HF model quantifying the influence of uncertain input parameters with respect to a predicted (uncertain) output.

2012 World Energy Outlook special report on unconventional gas exploitation concluded that the potential for contamination of surface and groundwater must be successfully addressed. The existence of both cons and pros for the HF method making it highly controversial. The European scene on SGD is totally split: there are a number of countries who seem to be quite enthusiastic, like Poland or the UK. Germany remains conservative and uncertain while France is skeptical. Even as a favourable country, the UK has undergone changes in its position. Its government announced in December 2012 to resume HF activities after a temporary moratorium following small seismic tremors. Research on HF is often funded by these government bodies. The debates on the HF induced consequences require a clear look and consultation where scientific and effective modelling tools predicting HF consequences is in urgent need. European Commission is making plans for developing a legal framework for shale gas exploration in Europe. The research methods developed by this project are expected to provide scientific modelling results supporting decision making in future energy development.
Figure 1
Figure 3
Figure 2
Figure 4