Summary Back production of proppant from hydraulically fractured wells, particularly those completed in the northern European Rotliegend formation, is a major operational problem, necessitating costly and manpower-intensive surface-handling procedures. Further, the development of unmanned platform operations offshore, required in today's economic climate, is impossible as long as this problem remains unsolved. The most cost-effective potential solution to this problem is provided by curable resin-coated proppant (RCP), which consolidates in the fracture. Early field trials with RCP's, however, were not completely effective in stopping the back production of proppant. Typically, some 10% of the total volume of RCP placed in the fracture was backproduced. Two types of RCP back production were identified: during well cleanup (Type A) and after a long period of proppant-free production (Type B). Type A is believed to be caused by an insufficient strength buildup of the RCP pack. The influence of factors affecting RCP pack strength buildup-resin type, reservoir (curing) temperature, resin/fracturing-fluid interaction (under shear and temperature), and erosion of the resin from the proppant grains, which can reduce the RCP pack strength-have been studied in the laboratory. Type B proppant back production was suspected to be caused by a previously unobserved phenomenon: damage resulting from stress cycling that the proppant pack undergoes each time the well is shut in and put back on production. Further, the applied stress increases as the drawdown is increased and the formation is depleted. We performed a laboratory study to help clarify the effect of curing temperature, water production rate, proppant size, and stress cycling on the integrity of RCP packs. The experiments confirmed the field experience that stress cycling has a dramatic effect on proppant back production of commercial RCP packs. The number of applied stress cycles (i.e., the number of times the well is shut in) and the initial RCP pack strength appear to be the dominant factors that govern proppant back production. Dedicated experiments are therefore required to evaluate the use of RCP's to eliminate proppant back production for a particular field application. Introduction Sand production is an operational problem that has plagued oil and gas wells producing from clastic formations since the early days of the oil industry. By contrast, proppant back production is found only in wells where hydraulically created fractures have been packed with (large) volumes of proppant. The proppant pack is unrestrained at the fracture mouth; once proppant grains enter the wellbore, they can be brought to surface with the well fluids. Such back production of proppant from hydraulically fractured wells, particularly those completed in the northern European Rotliegend formation, is a major operational problem. It necessitates costly and manpower-intensive surface-handling procedures (viz., the daily dumping of proppant) and on-site control of the chokes when beaning up the wells. Further, erosion of well and surface facilities presents a safety hazard, and proppant remaining in the wellbore can shut off production by covering the productive interval. Consequently, the development of unmanned platform operations offshore, required in today's economic climate, is impossible as long as significant proppant back production occurs. Incidentally, a similar tendency for hydraulically fractured wells to backproduce proppant is observed in Alaskan operations; however, owing to the different conditions (onshore oil production), the approach adopted there is "to live with it."