Deformation of a borehole across a salt bed in the U.S. Gulf Coast was monitored in support of a well cementing design. The well was drilled using a drilling mud of 13.3 lb/gal density through a salt-anhydrite section situated at depths of 17560 to 17870 feet. With the mud in place, caliper runs were made to measure the change in borehole diameter with time. The mud density in the borehole was then increased subsequently and caliper runs repeated until no noticeable change in borehole diameter within a 30-hour duration was recorded. When the mud density was 17.3 lb/gal, the borehole finally stabilized. Induced stress around the borehole was calculated. A relationship between induced stress (or mud density) and change in borehole diameter was established. Optimum density of cement slurry for the well was obtained. Results of the caliper measurements, along with the field measurement procedures are presented. procedures are presented Introduction Closure of deep wells drilled through salt in the U.S. Gulf Coast leading to the total well collapse is a known problem. Ideally, knowing the range of drilling fluid density that could be utilized for a particular depth-salt formation without causing particular depth-salt formation without causing noticeable deformation of the borehole would represent a positive contribution to drilling and subsequent well cementing technology. During the drilling operation, the mud column exerts a fluid pressure on the borehole wall, while the undisturbed pressure on the borehole wall, while the undisturbed salt bed is subjected to an interstitial fluid pressure and a natural state of field stress. pressure and a natural state of field stress. Thus, the reaction of these stresses induce a differential stress around the borehole. Assuming the state of stress in the undisturbed salt bed to be hydrostatic, which is approximately equal to the overburden pressure, the magnitude of induced differential stress around the borehole can be obtained by knowing the density of borehole fluid. As the drilling penetrates through salt, the state of induced stress in the gait bed immediately adjacent to the borehole may be of sufficient magnitude to cause plastic flow of salt, resulting in borehole deformation. If deformation results in a noticeable contraction of the borehole, the motion of the drill bit may be restricted or the emplacement of the casing after drilling ceases could be hampered. If, however, deformation results in a large expansion of the borehole, the salt bed may fracture and result in lost circulation. Insitu salt deformation around the borehole associated with the induced differential stress can be obtained by monitoring the diameter change of the borehole with time. A field measurement of the borehole closure was performed to determine in-situ differential stress performed to determine in-situ differential stress in support of the well cementing design. A well was drilled through the salt bed. The borehole was cleaned and left filled with drilling fluid. A series of caliper run was made. The mud density was then increased; thereby decreasing the stress difference. Caliper runs were repeated at the predetermined intervals for the same mud. The predetermined intervals for the same mud. The procedure was repeated until no noticeable change in procedure was repeated until no noticeable change in the borehole diameter was recorded. P. 133

SPE Members Abstract As an increasing number of gas fields are being developed or are planned to be developed in the permian red beds (Rotliegendes) of northwest Europe, hydraulic fracturing has become a major stimulation method to obtain economic production rates from this tight gas formation. Optimization of stimulation design and its implementation in the field become more crucial in the formation such as Rotliegendes, where a water bearing zone is underlain. The objective of this work was to simulate fracture growth under various conditions so that optimal treatments can be designed and implemented successfully to this tight gas reservoir. Formation rock properties of the payzone and adjoining barriers, fracturing fluid properties and their leakoff coefficients selected were experimentally determined in the laboratory using various Rotliegendes core samples. A 3-D model of Meyer and Associates was used to simulate the fracture length, width, height, proppant transport and settlement, fracture closure, and post-frac performance. The same treatment data were simulated using a 3-D model developed by Palmer and Carroll. Results obtained from the two models are compared on a rational basis. Two field applications are presented to demonstrate the proper design and successful stimulation treatments in the Rotliegendes. Introduction Since the discovery of the giant Dutch Groningen gas field in the north of the Netherlands (discovery well, Slochteren 1, 1959), an increasing number of gas fields are being developed or are planned to be developed in the permian basin of the North Sea, the Netherlands, Denmark, and Western Germany. The permian red beds of northwest Europe (or better known as "Rotliegendes" in Germany) are continental elastic sediments deposited under desert and semi-desert conditions. The mineralogical study of several Rotliegende core samples indicate that the sandstone primarily consists of quartz with minor dolomite, illite, and kaolinite (migrating type clays) and dawsonite (a bladed mineral). The general texture of such rock is shown in Figure 1. The proved and probable reserves of Rotliegendes gas in the North Sea, Netherlands, and Western Germany is estimated to be about 85x 1012 cu.ft. The development of these fields is only possible if the gas production rates can be maximized by effective hydraulic fracture treatments. Optimum fracture treatment techniques can substantially improve the supply of natural gas from these tight gas reservoirs. The effectiveness of hydraulic fracturing in a tight gas reservoir is strongly affected by the geometry of the created fracture. The geometry of a fracture deals with the propped fracture height and areal extent. The intrusion of a fracture from the payzone into the formations lying above and below is a serious concern in a fracture design. Moreover, if the hydraulic fracture is not contained within the producing formation and propagates in both the vertical and lateral directions, failure of the treatment can occur because there is a substantial loss of fracture fluid and proppant used to fracture the unproductive formations. The containment of a fracture within the producing zone is even more important where the underlying zone is water bearing. P. 247^

Once a reservoir is hydraulically fractured, the fracture may experience several repeated production/shut-in cycles. Evidence of reduced rate of recovery as wells are placed back on production, has been noted in gas wells that have been exposed to this type of cycling. As the cycling process is repeated, the propped fracture undergoes the ups and downs of closure stress, resulting in a gradual reduction in fracture conductivity. To observe the phenomenon, a propped fracture using formation rock, was subjected to cyclic loading in the laboratory. Change in fracture conductivity with time and the number of cycles was measured. Interaction between proppant and formation rock was examined. One sand size and two synthetic proppant sizes were evaluated. Two formation rock types of different hardnesses were used in simulating the fracture. At the same closure stress level, fracture conductivity continuously decreased as the loading cycle is repeated. The amount of reduction in conductivity depended upon the rock hardness and proppant size. For the same hardness rock, the magnitude of closure stress dictated the suitable proppant size to minimize the cyclic effect. Thus, the combination of information obtained through this study is a tool to select a proppant to minimize the cyclic effect. Suggestions and optimum design criteria to circumvent or minimize the repeated shut-in effect are presented along with the test results. Introduction For a hydraulic fracturing treatment to be successful, proppants must be placed across the producing zone to provide an effective propped fracture. Reservoir size and characteristics will determine the type and amount of proppant to be used. Properly placed proppant in the fracture will support the closure stress to maintain fracture conductivity high enough for transporting reservoir fluids to the wellbore. Surrounding tectonic stress being constant, the closure stress upon the proppant starts to increase as the well begins to flow, gradually deteriorating the conductivity of the propped fracture. This is considered to be a normal occurrence. However, the effective life of a propped fracture may be shortened by production methods. One such method is production cycling. Many fracture stimulated gas wells are subject to occasional shut-ins of short or extended duration due to demand restrictions or other reasons. Experience has indicated that the stabilized production rate, following the shut-in, is lower than the previous rate at the same wellbore pressure. An analysis of pressure buildup data indicated that fracture conductivity continued to decrease following the subsequent shut-ins. Repeated production and shut-in of the fracture stimulated well will cause cyclic loading onto the proppant in the fracture, thus, resulting in the decreased width of the propped fracture with cycles. To study this phenomenon, laboratory experiments were performed with a simulated propped fracture. Formatiion rocks of the Rotliegendes and the Wilcox sandstone were propped with Ottawa sand (20/40 mesh) and Intermediate strength proppants (20/40 and 12/20 mesh Interprop TM Plus) and subjected to cyclic TM- Interprop TM is a trademark of Norton-Alcoa Proppants. P. 351^