Wellbore Refluxing in Steam Injection Wells

Abstract

SPE Members Summary Most deep steam injection wells are completed with insulated tubulars to reduce heat loss. Due to various circumstances, the annular space between the tubing and casing is usually filled with water prior to the beginning of steam injection. Although most of the water is "boiled off" after injection commences, a small fraction of the water remains in the wellbore in the form of refluxing steam. Heat losses in a refluxing wellbore are considerably higher than would be expected if the annulus was dry. Refluxing was studied to determine the mechanisms which drive the process and to identify methods to eliminate refluxing. The analysis indicates that vaporization from bare surfaces such as subs and expansion joints located above the packer is the principle source of wellbore heat loss that drives the refluxing process. Although vaporization can occur when condensate contacts uninsulated couplings, such occurrences are transient and of short duration. Vaporization caused by heat losses from uninsulated couplings cannot sustain the process in the absence of vaporization from heated surfaces below. Therefore, insulating the couplings will not prevent wellbore refluxing. Other methods of controlling refluxing such as evacuation of the annulus, pressurization of the annulus with inert gas, and continuous injection of gas in the annulus were investigated. Both evacuation of the annulus and continuous injection of gas can prevent wellbore refluxing. Introduction The use of insulated tubing to reduce wellbore heat loss in steam injection wells began in the mid-1960's with the development of prestressed tubulars to prevent casing failure during steam stimulation of wells not designed to tolerate thermal stresses . In the last 15 years, insulated tubing has become widely used in steam injection wells spurred by the simultaneous growth of steamflooding projects as well as a significant rise in oil prices, and thus the cost of fuel to generate steam. In addition, there were technological developments in insulating materials that reduced the effective thermal conductivity of the insulating system from 0.03 Btu/hr/ft / F/ft to 0.01 Btu/hr/ft /F/ft. The evaluation of insulating tubing strings is usually done assuming the tubing string is run on a packer into a well with some standing water level. Water remaining in the annulus is assumed to be boiled off by heat transfer from uninsulated surfaces such as collars, subs, expansion joints and the packer body, leaving a dry annulus. Under these conditions, wellbore heat loss calculations show that heat losses from insulated tubing can be as low as 80–100 Btu/hr/ft. The economics of using insulated tubing strings are quite favorable with these heat loss rates even though the capital investment for the insulated tubing is significant. Recent field experience has shown that the annulus does not dry out in insulated steam injection wells. Data were presented demonstrating the existence of refluxing in the wellbore which maintained the casing temperature at a constant value consistent with the annulus pressure. Casing temperatures under refluxing conditions were maintained at 212F when the annulus pressure was 1 atm while the casing temperature in a dry annulus was expected to be about 130F. Because of this, heat losses were higher than anticipated which offset some of the economic benefits of using an insulated tubing string. P. 95^