TX 75083-3836, U.S.A., fax 01-972-952-9435. AbstractA low concentration chemical treatment has been developed for pumping into a producer from surface to stop water production without harming hydrocarbon production. Matrix reservoirs are the primary target, though the upper permeability limit is not yet known. Numerical simulation has shown that while such a treatment would not be universal, many viable targets exist. Horizontal wells and multilaterals can make ideal targets, as would many vertical wells in heterogeneous formations and hydraulically fractured wells. Key performance targets were set for these chemical selective water-blocking systems (often called relative permeability modifiers). A dilute, water-dispersed system was favoured as treatment volumes were expected to be large in some cases. Because some of the chemical will enter valued oil/gas zones, the chemical must collapse to allow oil/gas to flow back through treated rock. A mechanism for this is described. To be of wide use, the chemical system should inhibit water production in reservoir rock with matrix permeabilities up to 10 Darcy, and at temperatures up to at least 110 °C. Simulation indicated the minimum degree of water block required for selected targets. However, laboratory tests showed that blocks beyond 99.5% could be slow to clean up to hydrocarbon flow under available drawdown. This gave a performance target range. Well targets should always have appropriate reservoir stratigraphy, but it may not be possible to log the wells pre-treatment to check hydrocarbon/water inflow distribution. Therefore some "high-risk" treatments would fail because of lack of separate oil inflow, or absence of any flow barriers (such as shale) between water and remaining oil. A simple remediation treatment option was therefore required. Also, low toxicity and compatibility with production facilities were seen as non-negotiable properties. We describe the key molecular design parameters derived to give a polymer/crosslinker system with the performance demanded from simulation. Laboratory test results showing the required water/oil block performance and long term stability (at 100 °C) are presented. Field deployment designs including "rule of thumb" treatment volumes are described. The polymer and crosslinker were manufactured at full scale for trial application. Quality assurance tests of this unique "dancingthin-gel" system are detailed. Results of several full-scale "calibration" field trials are presented. Lab-field correlation appears to be acceptably robust.`