Formation damage traced to contaminated completion fluids

Aug. 12, 2002
Extensive compatibility tests determined the severity of the problem and developed preventive and remedial measures to deal with organic sludges that choked hydraulic fractures and the critical matrix in the producing reservoirs of the LL-652 oil field, in Venezuela.

Extensive compatibility tests determined the severity of the problem and developed preventive and remedial measures to deal with organic sludges that choked hydraulic fractures and the critical matrix in the producing reservoirs of the LL-652 oil field, in Venezuela.

Organic sludges form when iron-contaminated completion brines are lost to the formation. Although this was believed to be the primary cause of the problem, other plausible damage mechanisms include dehydrated polymer gel damage in proppant packs, fines migration, and water blocks.

LL-652 oil field

ChevronTexaco Overseas Petroleum Co. operates LL-652 oil field, in central-northeast Lake Maracaibo, Venezuela, under a 20-year operating service agreement, dated July 1997, with Petróleos de Venezuela SA (PDVSA). Its operations started in May 1998.

The field consists of fluvial deposits and shale-sand sequences that are compartmentalized by fault blocks, particularly in the north and along the west flank of the reservoir structure. The reservoir structure and sand quality decrease as one moves towards the north of the field.

Drill pipe and tubing rusts quickly in racks in the hot, humid Venezuelan environment (Fig. 1).
Click here to enlarge image

The reservoir had an initial water saturation that ranged from about 30% (immobile water) to as high as 45% in the north part of the field.

Most of the field has depleted reservoir pressure of about 1,500 psi, although pressure is near virgin, about 3,600 psi, in the north end of the field.

Reservoir depth ranges from 7,000 to 10,000 ft, and reservoir bottomhole static temperature (BHST) is 200-2400° F. Reservoir permeability ranges from 0.1-100 md with an average permeability of 10-20 md.

Most wells in the field have 100-300 ft of perforations distributed across as much as 1,000 ft of gross interval, and a few wells have been hydraulically fractured.

Recent observations indicated reduced well productivity because of some formation-damage mechanism.

Process mapping the completion and workover procedures revealed that contaminated completion brine lost to the formation likely was responsible for poor production rates. It was observed that clean brines become contaminated unintentionally when they come in contact with untreated tubulars that have corroded during exposure to air and rain on surface racks (Fig. 1).

Completions, workovers

The field used three general completion and workover processes (see accompanying completion and workover procedure box).

Click here to enlarge image

Figs. 2a-c illustrate conditions when contaminated completion brine losses may occur during each process.

Fig. 2a shows that losses will occur during a new well completion process if the well is perforated over balanced. Post perforating losses can obviously plug perforations with solids, and the filtrate from the losses may contain constituents that induce formation damage in the critical matrix.

The hole always will lose completion brine into the newly created propped fracture when sand is being circulated out of the wellbore (Fig. 2b). Such losses can be very damaging because of the possibility of creating a choked fracture, which is a fracture that is in poor communication with the wellbore.

Completion brine losses are less common during a major rig workover on a hydraulically fractured well because the procedure requires fluid-loss materials to be injected into old perforations, thus preventing formation damage. But, those plugged perforations must be cleaned after the treatment.

Workover procedures not involving hydraulic fracturing do not require fluid-loss materials, and under these conditions (Fig. 2c) some contaminated brine can be lost from the annulus into perforations above the packer element.

A literature review shows that brine losses can cause several types of formation damage including scale precipitation, clay induced damage, damage due to solids invasion, water block, emulsions, and sludge.1-12

Laboratory test procedures

One can evaluate the impact of scale, solids invasion, and water blocks without studying these mechanisms in a laboratory. But, laboratory studies must be conducted on clay sensitivity and brine-oil compatibility to understand their impact on the producing system.

Also, core tests illustrate water shock (clay sensitivity) and critical velocity for fines migration as well as brine-oil compatibility tests (see accompanying core flow procedure box).

The laboratory tests recreate downhole conditions to identify plausible formation damage mechanisms.

Brine-oil compatibility

The laboratory test used the following procedure to determine brine-oil compatibility:

  • Emulsion and sludge tendencies of the brine with produced fluids followed a Modified API Recommended Practice 42.13 14
  • A recommended modification to the sludge test filters the bottle contents through a No. 41 Whatman filter with a vacuum filtration apparatus.
  • Desirable characteristics include:
  1. Absolutely no sludge can be retained on the filter paper.
  2. Complete phase separation should be within 30 min at BHT, preferably within 5 min, and more preferably within 1 min.
  3. A sharp free-flowing interface should be observed between the acid and oil.
  4. The acid should strip the oil from the glass walls leaving the surface water wet.
  • Tests use wellsite brines containing all additives included in the brine formulation.
  • Tests use new medicine bottles, with the bottles discarded after each test. Nondisposal laboratory glassware used for emulsion tests were triple rinsed with alternating cycles of deionized water and 7.5% HCl.

Core-flow tests

The core-flow test extracted hydrocarbons from cores by injecting toluene, methanol, and then water in series into the cores. Test brine saturated the cores immediately after hydrocarbon extraction and before placement of the prepared cores in a Hassler-type core holder under a confining (overburden) pressure of 2,000 psi and a backpressure of 1,000 psi.

All cores were tested at 200° F. (93° C.).

An ISCO 2350 pump moved fluids through the cores, and low and high range Rosemount differential-pressure transducers measured the pressure drop across the core.

Click here to enlarge image

Fig. 3 shows flow apparatus similar to that used in the tests.

The fluids were mixed and filtered through 0.45 μm filters. All brines contained the standard additives (such as biocide) as used during LL-652 well completions.

Results

From the comprehensive suite of brine-oil compatibility tests, the major finding was that brine contaminated with ferric iron created massive quantities of sludge when mixed with crude oil from C-4, C-5, C-6, and Santa Barbara reservoirs.

All crude oils precipitated a moderate to a high mass of sludge when the brine was contaminated with only 100 ppm ferric iron, and some crude oils precipitated moderate amounts of sludge with only 20 ppm ferric iron contamination.

Ferric hydroxide precipitates in water contaminated with ferric iron. (fig 4a-4d)
Click here to enlarge image

Figs. 4-6 illustrate representative compatibility tests. Ferric hydroxide precipitates from iron-contaminated completion brine when the pH exceeds a value of about 2.

Fig. 4a shows the appearance of completion brine contaminated with various amounts of ferric iron. Asphaltenes are destabilized and precipitate from the crude when it is mixed with iron-contaminated brine. Fig. 4b shows the thick emulsion pad, stabilized by asphaltenes, that forms at the oil-brine interface.

Emulsion and sludge problems increase as the level of iron contamination in the brine increases (Fig. 4c).

Fig. 4d further illustrates the impact of iron on the creation of iron sludge. The bottle on the left has no iron contamination. The oil and water are separated by a clean interface, the interface is free flowing, and the interior glass surface is clean and water wet.

Sludge from a mixture of contaminated brine and crude forms on a filter (fig. 5)
Click here to enlarge image

The bottle on the right contains brine contaminated with 1,000 ppm Fe3+. The sludge that formed in the aqueous phase is so thick that it is self supporting. Fig 5 shows this sludge on a filter. The mass is thick and required a long time to separate from the water under vacuum filtration.

A short series of tests isolated the effect of brine additives on the brine-oil compatibility (Fig. 6). The iron-contaminated freshwater created a small emulsion pad when L58 was added to reduce the ferric to ferrous iron. Addition of Oxban oxygen scavenger (and Tetracide G biocide) to the freshwater containing iron and L58 sodium erythorbate reducing agent increased the emulsion pad volume.

All samples are freshwater contaminated with 1,000 ppm ferric iron. (fig. 6)
Click here to enlarge image

Failure to reduce the ferric iron will result in a precipitate (third cylinder from the left in Fig. 6) that may be an iron-sulfur compound.

Demulsifiers have an impact on the compatibility between brine and oil. (fig 7)
Click here to enlarge image

Another short series of tests identified potential demulsifiers to improve the compatibility between the contaminated completion brine and the crude oil (Fig. 7). The demulsifier provides good water wetting, clean emulsion break in iron-contaminated brines with and without ferric iron.

Core-flow tests

Two types of water-sensitivity tests were performed: water shock and critical velocity tests.

Water-shock tests identify brines that destabilize clays and fines when introduced into the pore space. Critical-velocity tests determine the interstitial velocity required to mobilize fines and clays. Core permeability should decrease suddenly above the critical velocity for fines migration.

Click here to enlarge image

Table 1 summarizes the mineralogy of the samples and Table 2 summarizes the critical velocity tests. From Table 2, one can see that no substantial mobilization of fines occurs at interstitial velocities of up to 10.4 cm/min for NaCl brines. This velocity is equivalent to about 1,440 b/d through a 100-ft openhole interval with 16% porosity.

Click here to enlarge image

The critical velocity for fines migration for 2% KCl brine appears to be about 4.2 cm/min, and for 9.4 ppg KCl brine, about 2.1 cm/min.

Cores exhibited a range of permeability losses during the water shock tests. A 2% NaCl brine caused no damage to the core. The other brines caused between 49 and 72% loss of permeability.

In every case, most permeability was restored during the reverse flow of lake water through the core.

Click here to enlarge image

Table 3 summarizes the water-shock results, indicating that these cores have moderate to low water sensitivity, depending on the brine tested.

The test results indicated a critical velocity or fines migration of 2.1-4.2 cm/min for KCl brines. The other brines did not mobilize any fines in the LL-652 cores under the critical velocity conditions tested.

The most compatible brine is 2% NaCl, whereas the most damaging brine is 9.4 ppg NaCl brine.

All cores recovered at least 73% of their initial permeability during a reverse flow test following the water shock experiment, regardless of brine used.

Recommendations

The tests determined that the most likely source of formation damage in the LL-652 wells was sludge formed by mixing iron-contaminated completion brines and crude oil. A secondary formation damage mechanism may be fines migration.

These conclusions led to the recommendation to avoid completion brine losses during the completion and workover process and to treat the completion brine with a combination of an iron-reducing additive and a demulsifier to minimize creating sludge and emulsions if losses occur.

Click here to enlarge image

The field adopted some of these changes during the second half of 2000. Figs. 8a and 8b show the benefits of these changes.

Wells completed using the original well completion and workover process (Fig. 8a) have an average initial productivity index (PI) of 0.00194/ft.

By comparison, wells (Fig. 8b) completed with the improved completion procedure had an average initial PI of 0.0042/ft, a 217% improvement from the older completion process.

An estimated 1,000-psi drawdown and 100 ft of net pay was used to determine the PIs.

Acknowledgments

Special thanks go to Petróleos de Venezuela, SA (PDVSA) and ChevronTexaco Overseas Petroleum Co. for giving permission to publish this article. The contributions of Schlumberger's Las Morochas and Sugar Land Client Support laboratories' technical staff are gratefully acknowledged.

Thanks also go to Oliver Villalba for his efforts in determining the performance of the wells completed prior to and after this study.

References

  1. Sharma, M.M., and Yortsos, Y.C., "Permeability Impairment Due to Fines Migration in Sandstones," SPE Paper No. 14819, International Symposium on Formation Damage Control, Lafayette, La., Feb. 26-27, 1986.
  2. Sharma, M.M., Yortsos, Y.C., and Handy, L.L., "Release and Deposition of Clays in Sandstones," SPE Paper No. 13562, International Symposium on Oilfield Chemistry, Phoenix, Ariz., Apr. 9-11, 1985.
  3. Vaidya, R.N., and Fogler, H.S., "Fines Migration and Formation Damage, Influence of pH and Ion Exchange," SPE Production Engineering, November 1992, pp. 325-30.
  4. Gruesbeck, C., Collins, R.E., "Entrainment and Deposition of Fine Particles in Porous Media," SPE Journal, December 1982, pp. 847-56.
  5. Browning, F.H., and Fogler, H.S., "Precipitation and Dissolution of Calcium Phosphonates for the Enhancement of Squeeze Lifetimes," SPE Production and Facilities, August 1995, pp. 144-50.
  6. Crowe, C., McConnell, S.B., Hinkel, J.J., and Chapman, K., "Scale Inhibition in Wellbores," SPE Paper No. 27996, University of Tulsa Centennial Petroleum Engineering Symposium, Tulsa, Aug. 29-31, 1994.
  7. Dubin, L., "The Effect of Organophosphorus Compounds and Polymers on CaCO3 Crystal Morphology," Journal of the Cooling Tower Institute, Vol. 3, No. 1, 1982, pp. 17-25..
  8. Foxenberg, W. E., Ali, S. A., Ke, M., Shelby, D. C., Burman J.W., "Eliminate Formation Damage Caused by High Density Completion Fluid - Crude Oil Emulsion," SPE Paper No. 39444, International Symposium on Formation Damage Control, Lafayette, La., Feb. 18-19, 1998.
  9. Figueroa-Ortiz, V., Cazares-Robles, F., and Fragachan, F.E., "Controlling Organic Deposits and Sludge in a Severe Hostile Environment," SPE Paper No. 31124, International Symposium on Formation Damage Control, Lafayette, La., Feb. 14-15, 1996.
  10. Loewen, K., Chan, K. S., Fraser, M., Leuty, B., "A Well Stimulation Acid Tube Clean Methodology," Petroleum Society of CIM/Society of Petroleum Engineers, Paper No. CIM/SPE 90-47, 1990.
  1. Alvarado, D.A., Marsden, S.S., "Flow of Oil-in Water Emulsins Through Tubes and Porous Media," SPE Paper No. 5859, 46th SPE Annual California Regional Meeting, Long Beach, Calif., Apr. 8-9, 1976.
  2. Kokal, S.L., Maini, B.B., and Woo, R., Flow of Emulsions in Porous Media, The American Chemical Society, 1992.
  3. API Recommended Practice for Laboratory Testing of Surface Active Agents for Well Stimulation, API RP 42, 1977.
  4. Ali, S.A., Shelby, D.C., Wagner, D.J., and Foxenberg, W.E., " New test identifies completion fluid compatibility problems," OGJ, Aug. 25, 1997, pp. 95-101.

Original well completion, workover process

Well completion process without fracture.

  1. Drill. Set 137/8-in. casing at around 1,200 ft, 95/8-in. casing at 5,500 ft (just below the B sands which are owned by PDVSA), a 7-in. liner will eventually be run to a TD of 7,500-9,000 ft.
  • Initial wells used Arco CF coring fluid (with 10,000 ppm Cl-, as NaCl).
  • Subsequent wells were drilled with Alplex mud, an aluminum complex for controlling sensitive shales.
  • Newest wells are drilled with Arco CF coring fluid (with 3% KCl).
  1. Condition the mud.
  2. Pump the chemical wash followed by MudPush, a viscosified pill, followed by cementing the 7-in. liner.
  3. Circulate to lake water.
  4. Use a bristle brush and bit and scraper to clean the interior surface of the casing.
  5. Chemical wash of the casing and drill pipe.
  6. Once everything is clean, fill the hole with 2% NaCl with filtered lake water or 9.4 ppg brine, or whatever density is required to control the well. Note, this brine is treated with Tetracide G biocide.
  7. Completion fluids are filtered to 2 μm. Iron contamination has been observed in the completion brines being circulated out. The 3% NH4Cl has appeared as a red-colored brine (iron hydroxide). The 2% NaCl brines have been observed to be yellow (ferric). Additionally, every rig adds 2 gal/100 bbl of biocide (Tetracide G, which is glutaraldehyde).
  8. The completion string is run in the hole with tubing-conveyed (TCP) guns (the completion string is not cleaned on the outside). The oxidized iron on the tubing exterior will contaminate the clean completion brine.
  9. Guns are fired underbalanced and dropped.
  10. The well is immediately produced following the firing of the TCP guns.
  • Emulsion and or sludge damage may occur if contaminated brine is lost to the formation.
  • If clean brine is reverse circulated and clean brine is lost to the formation, water block and or fines migration may occur.

Well completion process with fracture.

  1. Follow Steps 1-8 of well-completion process without fracture.
  2. The TCP guns are lowered on a work string.
  3. Guns are fired underbalanced and dropped. The workstring not being cleaned on the outside contaminates the clean brine that fills the hole.
  4. Multistage fracs are executed. Frac stages are isolated with sand plugs (both ClearFRAC, a viscoelastic surfactant-based frac fluid, and WF, water-based frac fluid, gels have been used to suspend the sand).
  5. The frac packer is released and lowered after the initial shut-in pressure (ISIP) and pressure fall-off are observed. The wells are then reverse circulated to remove sand plugs from the wellbore. There are appreciable completion brine losses to the formation, and this completion brine most likely contains junk including iron.
  6. Post-frac flowback
  • ISIP, observe pressure decline for a few hours
  • On last well only, the treating company spotted gel breaker (U66 mutual solvent in case of ClearFRAC, oxidizer or enzyme in case of guar).
  • Release packer and lower tubing to just above sand plugs so that sand can be fluidized out of the wellbore.
  • Reverse circulate the wells to remove sand plugs from the wellbore. There are appreciable completion brine losses to the formation and most likely, this completion brine contains junk including iron that was originally on the outside of the workstring.
  • Pull out of hole with the workstring.
  • Run in hole with production string, gas lift mandrels, packer, etc. The outside of new completion tubing is not cleaned.

Note: The entire completion process results in a long post-frac shut-in, minimum of 6-14 days. This shut-in time needs to be reduced to less than 24 hr if possible to take advantage of the artificially high pressure following the frac and to avoid extreme frac-fluid dehydration.

Major rig workover process.

  1. Kill well with 2% NaCl in filtered lake water containing biocide.
  2. Release packer and pull the completion. Packer fluid may contain 5 gal/100 bbl Tetrahib corrosion inhibitor, Tetracide G biocide 2 gal/100 bbl, and 2 gal/100 bbl Tetra Oxban oxygen scavenger.
  3. Run in hole with mill to clean the portion of the pipe where the packer will be set. Earlier wells were cleaned over the entire length of tubing and casing using well wash chemicals and bristle brushes. This cleaning process, however, was found to cause serious skin damage as the debris was carried into the formation with the kill fluid that was going south. Therefore, casing and outside of the workstring are not cleaned.
  4. Set the packer.
  5. Pump a fluid loss pill (millcarb) CaCO3 in brine plus XCD xanthan polymer until a squeeze is achieved if the well is to be fractured.Before fracturing, pump MaxSeal (ground K-max cross-linked polymer) down the annulus if there are severe losses to open perforations. These materials are used to balance the tubing pressure which may have a 10 ppg slurry in it if the fracture screened out prematurely.
  • Fluid-loss additives are normally used to keep old perforated wells under control. The CaCO3 and MaxSeal are particularly useful for preventing losses during the reverse circulation of sand plugs out of the well.
  • The CaCO3 and MaxSeal are not used if the well is not fracture stimulated. Instead the brine company keeps monitoring the fluid level in the annulus and will top it off with filtered 2% NaCl to control the well.
  • The same fluid losses will occur during reverse circulation if sand plugs are used to isolate the frac stages.

Core flow procedures

Critical-velocity tests

    Cut open wax seal and remove aluminum foil from core.
  • Prepare test brine.
  • Wipe oil off core exterior with a towel saturated in toluene.
  • Record the core dimensions and mass.
  • Load the core into the core flow apparatus (cut end at the effluent end).
  • Flush core plumbing with toluene.
  • Apply the overburden (2,000 psi).
  • Initiate flow of toluene at 0.25 cc/min. Apply 1,000 psi backpressure after one pore volume (estimate pore volume using a specified porosity).
  • Continue flushing core at 0.25 cc/min until effluent is clear.

  • Switch to injection of methan ol. Continue flushing the core for 10 pore volumes or until the effluent is clear.
  • Switch to injection of filtered test brine (lake water, 9.4 ppg KCl water, 2% KCl water, 9.4 ppg NaCl water, and 2% NaCl water).
  • Heat the cell to operating temperature (200° F.).
  • Measure initial, stable permeability once the pure test brine is flowing through the core (lake water, 9.4 ppg KCl water, 2% KCl water, 9.4 ppg NaCl water, and 2% NaCl water).
  • Increase the rate in a stepwise fashion: maintain the injection rate until the pressure drop across the core stabilizes.
  • Repeat until no more permeability decreases occur, or until a 95% perm reduction is achieved.
  • Reverse flow at 1 cc/min until stable permeability is achieved.
  • Remove 1/4-1/2 in. of the core plug effluent end for XRD, X-ray diffraction, analysis (minimum of 1 g of material). Use an oil based lubricant for the diamond saw.

Water-shock tests

  • Cut open wax seal and remove aluminum foil from core.
  • Prepare test brine and lake water.
  • Wipe oil off core exterior with a towel saturated in toluene.
  • Record the core dimensions and mass.
  • Load the core into the core flow apparatus (cut end at the effluent end).
  • Flush core plumbing with toluene.
  • Apply the overburden (2,000 psi).
  • Initiate flow of toluene at 1.0 cc/min (or less than critical velocity). Apply 1,000-psi backpressure after one pore volume (estimate pore volume using a specified porosity).
  • Continue flushing core at 1.0 cc/min until effluent is clear.
  • Switch to injection of methanol. Continue flushing the core for 10 pore volumes, or until the effluent is clear.
  • Switch to injection of filtered lake water.
  • Heat the cell to operating temperature (200° F.).
  • Measure initial, stable permeability (at two different flow rates, both of which are below the critical velocity).
  • Circulate the new test brine to the core-face (9.4 ppg KCl water, 2% KCl water, 9.4 ppg NaCl water, and 2% NaCl water).
  • Inject the test brine into the core at 1.0 cc/min.
  • Maintain rate until the pressure differential across the core stabilizes.
  • Circulate filtered lake water to the core-face. Inject lake water into the core at 1 ml/min. Record the stabilized permeability.
  • Reverse the flow direction and record the transient permeability.

Remove 1/4-1/2 in. of the core plug effluent end for XRD analysis (minimum of 1 gram of material). Use an oil based lubricant for the diamond saw.

The authors

Click here to enlarge image

Syed A. Ali is a senior staff research scientist for ChevronTexaco E & P Technology Co., Houston. He specializes in sandstone acidizing, formation-damage control, rock-fluid interaction, mineralogy, and oil field chemistry. Ali holds an MS from Ohio State University and a PhD from Rensselaer Polytechnic Institute. He is a member of SPE.

Click here to enlarge image

John H. Moran was a completions specialist for ChevronTexaco Overseas Petroleum Co. before his retirement. He previously worked for Texaco Inc. and Conoco Inc. For the past 2 years, he has worked as a consultant in China on Bohai Bay completion solutions.

Click here to enlarge image

Matthew J. Miller manages the stimulation client support lab for Schlumberger's North and South America group in Sugar Land, Tex. He works in matrix acidizing, acid and proppant fracturing, foam stimulation fluids, and mineral scale removal. Miller holds a BS in chemical engineering from the University of New Hampshire, and a PhD in chemical engineering from the University of Michigan.