2.2.3.I. Pulse injection

There are two main procedures for tracer pulse injection:

(i) Integral (topside) injection at the well head where the tracer enters all available perforated zones and is injected into the reservoir according to the injectivity in the various zones.

(ii) Downhole injection where different tracers may be injected in different isolated zones.

Topside radiotracer bypass injection: This is the simplest, cheapest and most frequently used injection method. The practical implementation procedure depends on the tracer to be injected, i. e. the procedure is simplest for beta emitters and somewhat more cumbersome for strong, high energy gamma emitters. The techniques range from mechanical crushing of tracer-containing glass vials in the injection stream to controlled pump operated injection and soluble solid state tracer slug injection.

Below, a well-proven technique for water radiotracer injection is described. The tracer mixture is prepared in a 100 mL flow through high pressure (rated to 500 bar) steel cylinder fitted with high pressure valves at both ends (total liquid volume ~ 70 mL). For beta tracers, a small quantity of a short lived gamm emitter (often 131I) is added for monitoring purposes.

Figure 7 illustrates typical small-sized injection equipment fitted into a suitcase shaped transportation container which may be carried by hand. The equipment is coupled to the main injection line as a bypass at two positions across a throttle valve. The function of the throttle valve is to set up a pressure difference between the two positions. The tracer container is connected to the equipment as shown. There are possibilities for pump operation of preparative fluids (inlet for chemicals) before tracer injection. The tracer is injected by by passed injection water driven by the pressure difference, and without any use of pumps. The injection efficiency is monitored by external gamma detectors in cases where a small quantity of a gamma emitting tracer is added to the tracer container. A typical injection time is <10 s for >99% of the tracer. However, rinsing continues

	Thrott e va ve
03 Valve 2
Manometer
Valve A( f
Valve D
Va ve C
gj Valve E
Va ve F
Valve I
Bottle valve G	Bottle valve H	Drain
Tracer bott e

Lead shield

FIG. 7. Small-sized injection apparatus for topside pulse injection of radiotracers.

for 60 min to clean out any remaining traces of activity from the injection apparatus.

For some tracers, it is advantageous to apply an extra non-radioactive molecular carrier in the injection phase (e. g. for 125Г and radiolabelled [Co(CN)6]3-). These tracers are injected using the same injection apparatus but now connected only to the main flow line at the outlet connection point. The injection is performed by pump operation where injection water containing carrier and other tracer preserving chemicals is pumped from an injection water reservoir.

Topside radiotracer ‘crushing’ injection: Figure 8 shows equipment for crushing injection of tracer. A 20 mL tracer glass vial containing a beta emitting tracer is loaded into the holder of the injector.

The equipment is then installed at the well head of the injector (Fig. 9). By operation of the hand wheels (A and B), the tracer vial will be crushed by the central bar and the tracer thereby released into the well. As with the bypass injection described above, a small quantity (a few megabecquerels) of a gamma emitting tracer, for instance 131I-, is utilized as a second tracer to monitor the injection process and confirm successful injection.

Pump operated radiotracer topside pulse injection: Pulse injection may also be carried out by means of pumping. When a gamma emitting tracer is used,

Connection tube 1

Connection

FIG. 8. Crushing injection equipment.

the radiation load to the operators has to be monitored in order to minimize any dose.

In this case, the transportation cage with the original lead shield may also be used during injection by leaving the radioactive liquid inside, connecting a slim flow line to the bottle and injecting the tracer by means of a high pressure suction pump. Useful equipment for this purpose is illustrated in Fig. 10.

Pump operated topside pulse injection of non-radioactive tracers: As mentioned previously, non-radioactive tracers are mainly liquid (aqueous) solutions of weak acids or salts. Over long distances, it is most convenient to transport the tracer compound in dry form (Fig. 11) and perform, whenever possible, the dissolution operation at the well site.

Some of the tracer compounds have limited water solubility. Weak acids may need addition of a base such as NaOH or KOH in order to promote dissolution. Finally, several hundred litres may be required for injection, even for an ordinary-sized reservoir section. Hence, injection requires higher capacity pumps than previously described. Figure 12 gives an example of both the tracer solution container and the pneumatically operated pumps used for injection.

This type of tracer injection may take a few hours depending on the volume of the tracer solution. In principle, this represents a square injection pulse but it may be regarded as an instantaneous pulse injection when compared with the transit time through the reservoir.

Downhole pulse injection: Downhole injection offers several advantages over topside injection:

• Removal of the danger for contamination of topside equipment. This may be especially important for radiolabelled [Co(CN)6]3- and possibly other complexes which may react chemically in the injection tubing.

• For stratified reservoirs, each zone may be uniquely labelled with a special tracer. This makes it possibile to examine vertical permeability in reservoirs and detect any extensive sealing.

• For horizontal wells which cover an extensive lateral reservoir section, zone injection is absolutely desirable for optimal information.

Downhole injection is not yet in general use although field tests have been successfully carried out [11]. A few attempts have been made to construct tools for general application, one of which is illustrated in Fig. 13 [10]. It is mainly constructed for vertical and deviated wells. It can be lowered into the well by a wireline which makes signal transfer possible. The tool is remotely operated from topside by PC control. It is based on the principle of a moving arm sealing onto the perforated section of the well through which the tracer solution is pumped at low speed during somewhat reduced rate of ordinary water injection. The tool is not yet in operation due to high cost of operation.

Lately, it has become technically possible to position downhole injection tools in horizontal wells by means of a well tractor. Combined with inflatable packers on the same line, sections may be isolated for specific tracer injection. Such injection equipment is composed of general and readily available components. It is therefore technically possible to conduct zone injection, but the first large scale field experiment has yet to be carried out. Some well completions are constructed to allow water injection into selected isolated zones. These completions can be used for selective zone tracer injection although the addition of tracer itself is carried out topside.

Both solvent extraction and ion exchange procedures may be used for entrapment, enrichment and separation of [Co(CN)6]3- and SCN- tracers from saline waters (sea water or formation water). It seems, however, that the ion exchange procedures are the least labour intensive.

The results reported indicate a higher solvent extraction yield for [Co(CN)6]3- with strong basic extractants rather than weak ones for the prevailing experimental conditions. This is valid both for the liquid extraction agents and for the solid anion exchangers tested. In the ion exchange processes, it is found that [Co(CN)6]3- is more strongly absorbed by Dowex 1 x 2 than by Dowex 2 x 8, although the absorption yields are close to 100% for both under the prevailing conditions. Their relative absorption strengths appear through their ability to release the tracer again by various elution agents.

The difference in the degree of cross-linking between the Dowex 1 x 2 and Dowex 2 x 8 resins would favour the kinetics of the x2 resin. However, for reasonable flow rates, the kinetics is not an issue. Since the Dowex 2 x 8 resin shows the highest elution yield (which was previously concluded to be due to the difference in the functional groups), use of the Dowex 2 x 8 resin is recommended for the actual separation process.

The remaining SCN — on the Dowex 2 x 8 ion exchanger after sample loading may be eluted selectively and with high elution efficiency by 10 mL of 0.1M NaClO4. This eluate may be added to the raffinate after the sample loading process and used as a feed to the BioRad AG1 column for separation and enrichment of the SCN- tracer along the lines already outlined.

From a 1000 mL sample, the total recovery (chemical yield) is >65%. Since the volume is reduced by a factor of 100, the enrichment factor is >65. By optimizing the process, it is probably possible to double this figure, especially by selecting carefully only the elution volume containing the bulk of the tracer peak. In this way, the volume may perhaps be reduced by a factor of 2-3 with a slight reduction in chemical yield.

The [Co(CN)6]3- complex may be stripped from Dowex 2 x 8 by, for instance, 10 mL 5M NH4NO3 (acceptable for modern scintillation cocktails) with an efficiency of 75%. The overall chemical yield will, therefore, be >75%. From the same sample size as for SCN-, the volume reduction will amount to a factor of 100 and the corresponding enrichment factor to >75. It is also possible to improve this figure by optimization of the procedure along the same lines as described for SCN-.

Appendix I

111.1.4.1. Well-trained personnel

The injections will be carried out by trained personnel from the tracer company. Most often, personnel from the oil or operator company are also present. As previously mentioned, two persons from the tracer company will participate in the injection programme. Both will have considerable experience in the technical application of radioactive tracers in industry and offshore activities. Before the injection work is started, a safety meeting will be held for the operater company’s crew where personnel from the tracer company will discuss their work and the safety aspects of the tracer injections.

111.1.4.2. Urine samples

Owing to the very low penetration power of beta radiation, the radiation from 3H, 14C or 35S intake cannot be registered by ordinary dose meters or radiation detectors. In order to be able to document possible radiation doses to personnel, or more likely, to document the absence of such doses in the case of tracer leakage, all operator company crew members that have to remain at the injection site will be asked to deliver urine samples. These samples will be analysed by liquid scintillation counting by a competent and independent health and safety laboratory. Analysis of urine samples is the most practical method for checking the intake of pure beta emitting radionuclides. Urine samples are taken before and after injection.

The interpretation of tracer experiments in interwell communications is much more complex than those obtained in industrial devices because oilfields and geothermal fields are non-boundary systems and little information about their internal structure is available; the curves are often incomplete, operating parameters change during the experiment and, finally, the percentage of tracer recovery is very low. In order to simplify the approach, an intercomparison of the software packages has been made on the experimental data obtained previously on a laboratory scale.

To achieve this purpose, the results from tracer experiments carried out in a laboratory model in the Commissariat a l’Energie Atomique in Grenoble, France, were used. To obtain the simulated tracer records the following conditions were considered:

• Water saturation (Sw = 1);

• Total layer thickness watered (h = 0.8 m);

• Reported porosity (0.35);

• Nominal water flow rates (injector: 107 mL/min, producers: 26.8 mL/min);

• Tracer recovery of 81.06% (in agreement with the reported values);

• No faults;

• No anisotropy;

• Dispersivity (estimated).

(i) Analysis of HTO

Water samples from oilfields may be of various qualities, ranging from relatively pure water (transparent liquid) via samples which contain some degree of oil to samples where the water exists mainly in an oil-water emulsion as illustrated in Fig. 23. Thus, pretreatment of the samples is needed before instrumental analysis of the beta emission (by liquid scintillation counting). The form of the pretreatment varies with the composition of the sample.

In general, for samples of type A and B, a combination of oil phase removal by pipetting and filtration followed by a distillation process is common. For samples of type C, an emulsion breaking step has to be included in the beginning before oil-water separation and eventual distillation of the resulting relatively small water volume (in most cases). Details on the analytical protocols, including pretreatments, are given in Appendix III.

(ii) Analysis of HTO and 14C labelled alcohol in mixture

The analysis of HTO and 14C labelled alcohol in water samples from the oilfield cannot be carried out directly by liquid scintillation counting, as purification somewhat different from the procedure described above for HTO is required. The following equipment parts and the purification procedure are proposed:

Equipment and reagents:

(1) Round flask 500 mL, connection size 24/40;

(2) Fractional distillation Vigreux column, 31 cm long, cs 24/40;

(3) Dean & Stark collector, volume 10 mL;

(4) Water cooling system;

(5) Heating mantle for the round flask;

(6) Magnetic stirrer;

(7) Liquid scintillation counter with standard vial of volume 22 mL;

(8) Methanol and toluene reagent grade;

(9) Scintillation cocktail, e. g. Instagel or Ultima Gold.

Procedure:

(1) Sample treatment: Add 10% v/v of toluene to a water aliquot in a separation funnel and shake to extract any dispersed oil droplets into the toluene.

(2) Transfer a maximum 10 mL aliquot of the water phase to a standard counting vial and mix with 10-12 mL of an appropriate scintillation cocktail. Shake vigorously to create a stable gel.

(3) Store the sample in the dark at least 1 h before starting the count.

(4) Count the sample in the liquid scintillation counter in dual label mode with the windows set at 3H and 14C. Calculate the tritium activity by correcting for the contribution of 14C in the tritium window.

(5) To another 10 mL of the purified water phase add 3 mL of methanol and 1.3 mL of toluene and transfer the mixture into the round flask. This volume ratio of methanol and toluene creates an azeotrope.

(6) Heat the sample very gently at low power (a few hours) to distil off the azeotrope.

(7) Collect the distillate (about 4 mL) into the vial.

(8) Add cocktail (10 mL) and count in dual label mode with the windows set at 3H and 14C. Calculate the 14C activity by correcting for the tritium activity in the 14C window.

The injection of tritium in the LGPF has indicated the applicability of tritium as tracer, as it has been detected in three wells. Its low recovery, however, could be due to insufficient concentration injected into well 1R8D. It could also indicate that there is significant dispersion or diffusion of tritium within the reservoir, such that its occurrence in other wells could occur at a much later date.

However, if insufficient concentration was injected, there is a possibility that the tritium injected will not find its way out of the reservoir.

The absence of processed historical matching (Fig. 76) in terms of thermal decline in the wells monitored for tracer could indicate that there are other sources of brine/cooler waters that effected a change in the wells. The tracer tests conducted for tritium in LGPF were able to detect the connection between the injector and the production wells monitored.

HTO with an activity of 10 Ci was injected into well 1R8D and tritium was detected at three monitor wells: 2R3D, 214 and 202.

These wells lie directly on the north-east path of the tritium derived from well 1R8D, along the Sambaloran Fault. The other monitor wells, which are situated to the north of the injector, did not manifest any tritium breakthrough one year after injection. The recovery in these wells, however, is only 0.1-0.4%. Near simultaneous NDS tracer test injected into the same well revealed positive breakthroughs, with recoveries slightly higher at 0.1-1.3%. Both tritium and NDS yielded the highest recovery at well 2R3D, the well nearest the injector.

The tracer results showed the hydrological connection between the wells monitored. The very low recovery, however, suggests other possible paths of the fluids from 1R8D. This was not established in the monitoring programme conducted. Nonetheless, the exercise demonstrated that tritium can indeed be utilized as a tracer in a vapour dominated environment. The major consideration here would be the cost of the tritium and the analysis.

111.3.1. Method

The tributylphosphate solvent extraction method was developed for 14C or 35S labelled SCN- enrichment. The recovery efficiency of S14CN — or 35SCN- is about 90%. The enriched sample is measured by liquid scintillation counting. Essentially, the method consists of three steps: [4] [5] [6]

1.1. PLANNING AN INTERWELL TRACER TEST

1.1.1. Wells to trace

The purpose of interwell tests and the kind of technical information that can be derived from them are described above both for oil fields and geothermal fields.

For both reservoir types the interwell tracer tests give quantitative information on the fluid dynamics in a reservoir. Dynamic information from a reservoir may, in addition, be obtained by three other methods: (i) logging of production rates (profiles) of reservoir fluids, (ii) pressure testing and (iii) time- lapse seismic examinations (4-D seismic). However, these methods and tracer testing are complementary and cannot directly replace one another.

Selection of wells to trace must be based on the best available information on the reservoir and the additional knowledge needed in order to optimize reservoir performance. The technical personnel with the most intimate knowledge of the reservoir lithology, stratigraphy and structure are the reservoir engineers together with reservoir geologists. Therefore, an optimum selection of wells should be performed in teamwork with these specialists and the tracer specialist.

Microbial stability shows up in the tests described above owing to a substantial degree of degradation at temperatures below 70°C, while the tracer

survivability may be better at higher temperatures. An example of this is the formate ion H14COO — that seems to be degraded to 14CO2.

It is also possible to add certain specific bacterial cultures to the vials to check the effect. The experiments may be somewhat difficult to control. Aerobic and anaerobic tests may be performed.

A laboratory intercomparison test was performed where the same set of data was provided by Argentina to coordinated research projects participants. Results from using the Anduril 2.3 software on these data are displayed in Figs 85-89.

11.3.3. Discussion

The values of the obtained parameters were very similar for all the users of the Anduril software.

However, a problem was detected in relation to the computation of the ‘final time’. The strong differences in the calculated cumulative tracer recovery of well K-301 indicated that an appropriate final time must be chosen.

It is opportune to highlight that an elemental moment analysis may be used to calculate temporal and volumetric moments. The use of analytical solutions allows the number of producing layers (by a deconvolution process) to be identified and, thereby, the moments (and the associate parameters) for each layer to be calculated.

Pike time (days)

400 300 200 100 0

Argentina Brazil France Indonesia Pakistan Philippines Vietnam China

Breakthrough time (days)

4
Argentina Brazil France Indonesia Pakistan Philippines Vietnam China

1500-и 1400 1300 1200 1100 1000 900 800 700 600 500 400 300 200 100 0

Argentina Brazil France Indonesia Pakistan Philippines Vietnam China

However, it should be noted that the analytical solutions from one dimensional models are less dispersive than the reality (which is two or three dimensional). Also, the one dimensional models cannot consider the ‘areal extension’ of the tracer flow (identical to the water flow). Consequently, very important parameters such as layer thickness and layer permeability cannot be calculated.

Appendix III