Cementing gas wells in Khuff reservoir where sustained annular pressure has been reported in many wells, presents a big challenge offshore in Umm Sheif field in Abu Dhabi. The main challenges are preventing gas migration and achieving long-term zonal isolation using a competent cement sealant system able to withstand downhole stresses and high temperatures during production cycles. An extensive study was carried out on two previously cemented wells, and an in-depth analysis was performed on the cement systems pumped in this field. The knowledge gained paved the way for designing, planning, and executing successful cementation on a recent well. A mathematical model simulating downhole stresses in terms of pressure and temperature was used in designing the cement sealant system. Unlike conventional cement systems, properties such as high Poisson's ratio and low Young's modulus value compared to that of the rock were optimized in the new system to achieve mechanical resistance and durability. This paper describes a case history whereby this cement sealant system was used to cement and secure the critical phases of this well. Cement evaluation logs across all the liners showed complete cement coverage and zonal isolation, especially across the Khuff gas reservoir. Introduction Any well containing a gas bearing formation is a potential candidate for annular gas migration. Cementing wells with potential gas migration is a critical operation. It requires from both service company and operator full and complete cooperation to gather the necessary data in order to optimize the design of the cement slurry, as well as its placement to mitigate the risk with the gas migration. The severity of gas channeling or migration can range from the most hazardous, blowouts, to less severe cases of residual gas pressure of a few psi at the wellhead. The main concern is the gas migration behind shallow casing strings or behind long liners For shallower casing strings cemented to surface the slurry volumes are usually large, holes are washed out, centralization is poor and frequently the fracture gradient will not allow the pumping of heavier fluids. For liners, more attention is usually paid to design and mud removal, however long liners pose problems of large temperature differentials between the top and bottom sections. This poses problems when the gas zone is near the top of the liner section.

The zonal isolation behind casing was critical during a major field development project in the Middle East. However, the assessment whether the cement bond will hold against hydraulic communication or not has not always been an easy task; slurry contamination during pumping, slurry invasion by formation hydrocarbons especially gas, channeling through the cement sheath, patchy cement, annulus shape and casing excentralization inside the outer wellbore are all factors that have traditionally made it difficult and sometimes even impossible to judge the zonal isolation behind casing between various zones of the reservoir. A new approach was adopted for the assessment of the cement bond behind casing in this project. The technique combines the conventional "pulse -- echo" ultrasonic measurement with that of a newly introduced flexural waveform. The analysis from this combination allows a better discrimination between solid, liquid and gas behind casing and it also provides, for the first time, a comprehensive assessment of the annulus shape and casing centralization inside open hole or inside the second casing string. This latter casing stand-off measurement has proven to be instrumental when deciding whether the cement sheath will hold against hydraulic communication or not. In this paper we will discuss the data that was acquired in four wells from this project. We aim at proving the clear benefit of this new approach for the assessment of zonal isolation behind casing. Introduction The traditional ultrasonic measurement estimates the acoustic impedance of the media behind the casing wall. The acoustic impedance of any material is the product of its density by the sound velocity through it (Z = density × acoustic velocity). Z is usually expressed in MRayl (106 kg.m-2.s-1). The readings are then binned in three "baskets" in order to determine what points denote gas, which ones indicate liquid and which ones indicate solid. Empirical cutoffs or thresholds are used to determine the extent of each basket. The default value of the threshold between liquid and solid is 2.6 Mrayls usually. The material at any point on the log where the acoustic impedance measures above this value is considered solid. hardened non-light weight cement acoustic impedance is usually distinctly higher than this value, whilst liquid and cement slurries have an acoustic impedance that is lower than 2.y Mrayls. There is no ambiguity in the classification of the material behind the casing in this case. However the acoustic impedance of hardened light weight cements can also be lower than this value (see Table 1). The contamination of the cement that might occur during the cement job can decrease its acoustic impedance below this threshold too. Consequently there is a range where acoustic impedances of slurries and heavy mud overlap with those of solid light weight and/or contaminated cement (Fig. 1). A "single axis" measurement of the acoustic impedance is not enough to decide on the nature of the material behind casing in this case. In the new technique, an additional measurement of the flexural attenuation has been added in order to solve this problem.