Simultaneous Use of Ester and CO2 Single-Well Tracers
- S.L. Wellington (Shell Development Co.)
- Document ID
- Society of Petroleum Engineers
- SPE Reservoir Engineering
- Publication Date
- November 1994
- Document Type
- Journal Paper
- 240 - 246
- 1994. Society of Petroleum Engineers
- 5.2.1 Phase Behavior and PVT Measurements, 4.1.2 Separation and Treating, 5.6.5 Tracers, 2.2.2 Perforating, 3.1.6 Gas Lift, 5.3.2 Multiphase Flow, 4.2.3 Materials and Corrosion, 4.1.5 Processing Equipment, 1.6.9 Coring, Fishing, 4.3.4 Scale, 5.5.2 Core Analysis, 2.4.3 Sand/Solids Control, 3 Production and Well Operations, 5.3.4 Reduction of Residual Oil Saturation, 5.1 Reservoir Characterisation, 5.6.4 Drillstem/Well Testing
- 0 in the last 30 days
- 226 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 5.00|
|SPE Non-Member Price:||USD 35.00|
This paper documents the consecutive and concurrent use of both ester andCO2 tracer chemistries in the same single-well test application. The simplicityof the CO2 system, which uses water-soluble-only injectants, and a no-soak/soaktest sequence allowed determination of definitive residual oil saturation (ROS)values.
The tests were applied in a California turbidite reservoir and in a U.S.gulf coast sandstone reservoir. In the turbidite, the CO2 test was run betweenthe mini- and main ester tests. In the sandstone, a separate CO2 test wasfollowed by injection of both reactive chemicals in the sameinjection/production sequence. Including both reactive chemistries in the sametest sequence appears to be a powerful test method to determine and to verifythe tracer responses and oil saturation. Ref. 1 describes the ester chemistryresults; documents the large ester hydrolysis pH-rate dependence; and concludesthat transit-generated tracer can dominate the response, making the testinsensitive to oil saturation. We redesigned the ester test to include ano-soak/soak test sequence to measure ester hydrolyzed in the reservoir duringtransit and soak periods, which indicates whether the tracer was generatedduring the soak period and therefore whether a unique oil saturation can becalculated in principle from the response.
Ref. 2 introduced the use of in-situ-generated CO2 as an oil tracer. Thefield procedures are essentially the same for ester and CO2 chemistries, butreactive and nonreactive tracers in the CO2 system travel together at thevelocity of the brine during injection because they are both whollywater-soluble. A wide range of robust and predictable CO2 generators withdifferent reaction rates is available. This allows optimal generator selectionfor a particular field application. These characteristics greatly simplify datainterpretation.
California Turbidite Tritiated-Water and CO2 Results
The ROS in a California turbidite reservoir was measured with the solubleCO2 tracer method. An ROS of 41 2 PV% was calculated with one layer andreversible flow. Fig. 1 compares the measured and calculated cumulative waterand oil tracer responses. The well was quite competent, and the water tracerarrived within 5 bbl of a perfectly reversible injection and production cyclefree of fluid drift. This can be seen readily by the match between the idealcalculation and the smooth "S"-shaped tritiated-water response.
More than 52,000 water and oil tracer data points were recorded.Approximately 35 points (at every 1,500th water and oil tracer data point) areplotted. The tritium and CO2 tracers were determined by use of acomputer-controlled on-line analysis system that continuously sampled aslip-stream of produced fluids and recorded the response on magnetic media.Tritium was determined by scintillation counting, and CO2 was determined bytriethanol amine titration.
The oil tracer response was good over most of the produced volume andallowed accurate calculation of the volume-averaged ROS. However, high drawdownproduction conditions, especially evident during the last 15% of production,caused gas and oil expulsion and therefore nonchromatographic tracerproduction. Oil production changes the transport mechanism by bringing the oiltracer to the wellbore directly instead of through the water phase only afterpartitioning. The nonchromatographic (fast) tracer transport is obvious in Fig.1 near the end of the test.
Although oil expulsion marred the appearance of the results, it fortunatelyoccurred late in the production cycle and did not interfere with oil saturationdetermination. Bottomhole pressure was 2,000 psi during shut-in and as low as80 psi during production. The bubblepoint pressure of the reservoir oil is1,000 psi. Oil desaturation by gas-bubble generation is a well-knownphenomenon. Separate simulations indicated that pressure drawdown below thebubblepoint would occur in the near-wellbore region ( 2 to 3 ft) under theseconditions. Chromatographic transport simulations placed the remaining tracerin the near-wellbore region and corroborated the observed fast tracertransport. The Appendix provides additional comments on the high drawdownconditions and the small amount of corrosion interference during the first twowellbore volumes of production.
The 41% CO2-tracer-test-determined ROS value agrees well with the 40%calculated from the ester results when the pH-dependent rate effects wereincluded in the analysis. Unfortunately, the ester value cannot be used forcorroboration because of excessive tracer generation during transit (see Ref.1).
ROS Calculation. The same governing chromatographic equations apply to bothtracer chemistries. Therefore, most hydrolysis must occur during the soakperiod. In comparison with the pH-dependent ester calculations that requiredifferent hydrolysis rate constants for the injection, soak, and productioncycles of the test, the same hydrolysis rate constant is used throughout forthe CO2 single-well tracer test (SWTT) calculations. This is appropriatebecause temperature was the only parameter found to affect the reaction ratefor most CO2 generators.
Temperature gradients are not a problem because water-soluble CO2 generatorstravel with the brine front and therefore are placed beyond theinjection-fluid-cooled region. The CO2 generator may be completely hydrolyzedduring the soak period without harming the test because it is independent fromthe water tracer, which is desirable. Complete hydrolysis during the soakperiod precludes reaction during production and simplifies interpretation.Essentially, this condition is achievable under most reservoir conditionsbecause many generators with different reaction rates are available.
Tracer Response. The water tracer response from the turbidite test is nearlyideal; therefore, the calculated oil tracer response is also nearly ideal. Theconstitutive relationship between the associated tracer-determined arrivalvolumes, Voi and Vwi, for calculating the oil saturation is
When the Voi/Vwi ratio is constant throughout tracer production, therequired quasistatic equilibrium condition for application of thechromatographic equation is verified and oil saturation is determined uniquelyfor a specified partition coefficient, Ko. Dispersion is the same for bothtracers because they return in the brine phase from the same position in thereservoir.
The calculated oil tracer response in Fig. 1 is actually a"predicted" response (i.e., predicted from the tritiated-water responsefor any assumed oil saturation by Eq. 1). For example, Fig. 2 shows measuredwater and oil tracer data and the predicted oil tracer responses from 0% to 60%ROS in 10% increments. Fig. 3 overplots the measured and predicted 39%, 41%,and 43% ROS values and defines the sensitivity of the results.
|File Size||657 KB||Number of Pages||7|