Quartz arenites are very clean quartz sandstones with unusually high permeabilities for any given porosity. This is attributed to the presence of large, clean pores which are often well connected. These large pores are susceptible to drilling damage, mainly from the invasion of solids. This paper presents both laboratory and field data on the development of drilling fluids to minimise damage in quartz arenites. The required changes in field mud engineering practices, to achieve improved well performance, are discussed. Field experiences are presented, including the impact of the modified mud systems and engineering practices on the resulting skin factors. Introduction Quartz arenites are very clean quartz sandstones with unusual poro-perm properties. Significant permeability is retained at low porosities and as the porosity increases the observed permeabilities are much higher than would normally be predicted. This unusually high permeability is attributed to the presence of large, clean pores which are often well connected. Despite these favourable characteristics, many wells drilled in quartz arenite reservoirs have not been as productive as expected, and have had high skin factors. In the absence of reactive clays, many recognised damage mechanisms are not applicable in quartz arenites, yet drilling damage is still seen to occur. Laboratory data shows that conventionally designed drilling fluids can cause high levels of damage, but that mud filtrates alone do not cause significant damage. The main damage mechanism is solids invasion into the large pores, and it is shown that changes in drilling fluid design can minimise this damage. These modifications to the drilling fluid require changes in field mud engineering practices, in order to achieve improved well performance. Field experiences are presented, including the impact of the modified mud systems and engineering practices on the resulting skin factors. The Mirador Formation Quartz arenites are clean quartz sandstones, which contain more than 95% quartz. The Mirador formation in the Cusiana field in Colombia is a quartz arenite, with a very low clay content. The typical mineralogy is 78% quartz grains, 14% quartz cement and 8% porosity. Despite low porosity, permeabilities are high. For example, 8% porosity corresponds to about 100mD and 10% porosity to about 4o0mD, although permeability is dramatically affected by grain size. The permeability of the very coarse grained sand is about 800mD at 10% porosity, whereas the permeability of the fine grained sand is about 90mD at the same porosity. The completed intervals have a wide range of permeability, sometimes from less than 10mD to more than 5000mD (the >1000mD sands generally have low thickness). The Mirador has very low vertical permeability, compared to horizontal (Kv/Kh = 0.1 or less, dependent on formation permeability). Reservoir pressure is approximately 5300psi and reservoir temperature is 127 F. Pore Size Distribution. The high permeability at low porosities is a result of large pore size and good interconnectivity. The median (D50) pore size in the lower Mirador is usually about 60, although in the high permeability streaks the D50 pore size has been measured as high as 110. More importantly, typical pore size distributions show the D95 to be substantially higher than the Dso (Fig. 1). As an example, a core with Dso pore size of 60 will have a D95 of 350-400. By converting a typical pore size distribution to a predicted permeability distribution (Fig. 2) it is apparent that the large pores dominate the permeability. In the example presented, the largest 30% of the pores account for about 85% of the permeability. Consequently, the large pores need to be protected in order to minimise damage. Compressive Strength. As a result of the secondary quartz cementation, which produced the low porosities, the Mirador sands also have high unconfined compressive strengths. P. 147

The productivity of wells damaged during drilling is directly dependent on the depth of the damage and the performance of the perforating guns. If the perforations by-pass the damaged zone then the well will have a low mechanical skin. Conversely, if the depth of damage is greater than the perforation length, the skin factor will be much higher, especially when the drilling damage is severe. While we normally associate drilling damage with low strength rocks, there are reported hard-rock fields with extensive drilling damage. The performance of shaped charges is significantly affected by the compressive strength of the rock to be perforated; consequently, the ability to bypass drilling damage in formations with high rock strength is reduced. Previously reported work has shown a 75% reduction in total target penetration, compared to API Section I, in rock with an unconfined compressive strength of approximately 25,000 psi. This paper describes the development and field testing of alternative charge designs aimed at improving performance in high compressive strength formations. So that the adverse effects of drilling damage can be reduced, computer simulations and laboratory tests showing the improvements achieved are presented. Field testing of the new charges and results achieved are shown. Introduction The basis for this project was to increase charge penetration depth to help optimize the completion efficiencies for the hard rock sandstone reservoirs in South America. Due to the unique properties of these quartz arenite sandstones, high compressive strengths up to 25,000 psi are common, and corresponding penetration depths are reduced. Additionally, these reservoirs have high permeability and modest porosities, resulting in large pore throats. Formation damage often occurs during the drilling of a well. Exposure to a drilling fluid generally results in the invasion of the rock matrix by mud filtrate and by mud solids. The extent of this invaded zone is dependent on several factors, such as the fluid loss characteristics of the mud system, the applied overbalance, the pore size distribution of the rock matrix and the time taken to drill the zone. The invaded zone may range from a few inches to a few feet around the well and usually results in a reduction of permeability. This permeability reduction, or damage, can have a dramatic impact on the potential productivity of the well. The most important consideration with respect to perforation length and well productivity is whether there is drilling damage and if the perforation length is sufficient to bypass such a damage zone. It is normally expected that effective perforating will bypass formation damage around a well if this damage is limited to a few inches. In hard rocks, the probability of bypassing the damaged zone is reduced due to the reduction in observed perforation length. For hard rocks with a significant depth of formation damage it is unlikely that the perforations will reach beyond the damaged zone. The effect of perforation length on well productivity has been reported by McDowell and Muskat, Harris and Klotz et al. These studies showed that the well productivity could only be maximized if the perforations penetrated beyond the damaged zone. Even when only a few perforations just pass the damaged zone the observed impact on productivity is significant. P. 33^

No abstract

Large, high density fracture networks are necessary to deliver commercial production rates from sub-microdarcy permeability organic-rich shale reservoirs. Operators have increased lateral length and fracture stages as the primary means to improve well performance and, more recently, are tailoring completion techniques to local experience and reservoir-specific learning. In particular, closer fracture stage spacing or increased number of stages per well have driven improvements in well performance. Large scale adoption occurs when the change in performance is clearly linked to the reservoir-specific completion design.Horizontal well fracturing efficiency in unconventional reservoirs is notoriously poor. Numerous authors report that 40 to 60 per cent of frac stages or individual perforation clusters have been shown (albeit with highly uncertain surveillance methods) to contribute little or no production. The fracture initiation and propagation process is very complex in shale; it is affected by in-situ stress, geomechanical heterogeneity, presence of natural fractures, and completion parameters. Close cluster spacing can provide enhanced well production; however, if the spacing is too close, stress shadowing among these clusters can actually induce higher stresses, creating fracture competition.This paper presents an approach to the integration of these parameters through both state-of-the-art geological characterization and unconventional 3D hydraulic fracture modeling. We couple stochastic discrete fracture network (DFN) models of in-situ natural fractures with a state-of-the art 3D unconventional fracture simulator. The modeled fracture geometry and associated conductivity is exported into a dynamic reservoir flow model, for production performance prediction. Calibrated toolkits and workflows, underpinned by integrated surveillance including distributed temperature and acoustic fiber optic sensing (DTS/DAS), are used to optimize horizontal well completions. A case study is presented which demonstrates the technical merits and economic benefits of using this multidisciplinary approach to completion optimization.

The productivity of wells with non-perforated completions can be impaired by both mud filtrate invasion and incomplete mud filter cake removal. Well inflow modelling illustrates that the percentage of the interval flowing and the distribution of the flowing intervals, are more important than the filtrate induced permeability reduction around the wellbore. Therefore adequate filter cake removal is essential for optimum well performance. This paper addresses both physical and chemical methods of mud filter cake removal.