NEW CORROSION INHIBITOR MORE UNIFORMLY PROTECTS DOWNHOLE OIL WELL SURFACES

R.L. Martin
Petrolite Corp.
St. Louis, Mo.

Neil B. Horn
Unocal Corp.
Snyder, Tex.

A new oil well batch corrosion treatment feeds downhole inhibitor more uniformly and can save costs by increasing treatment intervals and/or improving corrosion control.

This inhibitor differs from conventional inhibitors, not in chemistry, but in physical properties. Knowledge of corrosion and of corrosion inhibitor performance is continually being improved.1-4 With this new inhibitor, application methodology is similar to the batch-and-flush method used successfully for many years in oil fields.5 However, because the inhibitor is contained in a metastable emulsion, it possesses physical properties that allow a steadier inhibitor feed from the well annulus to corroding metal surfaces.

FIELD TESTS

Field tests illustrate how this new treatment can often reduce inhibitor cost and, in most cases, failure frequency. Bottom-line cost of corrosion protection can be critical to oil field economics.

TEST PROGRAM

Except for individual wells hand-treated early in the test program, the new inhibitor was introduced into wells with a treating truck. Flush brine came from the truck in all except a few early wells that were flushed with diverted production.

Preliminary data gathered for test wells included production rates, fluid levels, corrosion history, corrosion mechanism (sweet, sour, enhanced oil recovery, oxygen entry, etc.), and economics. Sonic methods determined fluid levels, 6 and corrosion rates were measured by coupons and linear polarization methods. 7 Brine chemistry measurements and potentio-dynamic polarization shed light on corrosion mechanisms. 8-10

In early tests, corrosion rate recording instruments provided information from which inhibitor feedback could be inferred. In a few cases, inhibitor residuals in the produced oil were measured with a high-pressure liquid chromatography (HPLC).

TEST RESULTS

Table 1 (12821 bytes) lists inhibitor performance characteristics in individual test wells. In these varied test wells, the new treatment usually gave low corrosion rates (measured by linear polarization). The rates were lower than seen with previous inhibitors even when previous inhibitors were optimized.

The lower minima are probably due to improved inhibitor film building conditions. More uniform feed-around provides a longer time for the inhibitor to incorporate into the corrosion product film. 11 Certain wells need specialized inhibitors for adequate control conventionally. In particular, this is in polymer and CO2 enhanced-oil-recovery projects. 8 Some shallow wells (Table 1 first entry) (12821 bytes) experienced corrosion influenced by small amounts of oxygen. Using conventional treatments, these wells would require special inhibitor chemistry. 7 9 The more favorable film building mentioned previously, is the likely reason that the new treatments were more effective.

Another significant observation is that the effective control following batch treatment is always longer. This held true even though test wells had sweet, sour, primary, secondary, or tertiary production.

Treatment intervals in Table 1 (12821 bytes) were calculated based on time required to exceed 1 mil/year (0.025 mm/year) after the treatment. In many of the wells treated conventionally, continuous corrosion inhibitor injection was used.

With the new methodology, wells previously requiring continuous treatment now usually received one batch/week. The likely reason this worked is because of the improved feedback profile (Fig. 1) (11031 bytes). These curves represent direct residual measurements and are confirmed by inference from recorded linear-polarization corrosion rates.

The profiles for conventional treatment in Fig. 1 (11031 bytes) would vary according to annular fluid level. Even sharper maxima would be seen in pumped-off wells treated with conventional inhibitors.

The profiles in Fig. 1 (11031 bytes) are from a well with 800 ft of annular fluid above the pump and producing about 800 b/d of total fluid. Both inhibitors are oil soluble so that the analyses were made in the oil but normalized to a concentration based on the entire sample.

The same batch size, 25 ppm based on 1 week's total production, and flush volume went into the well for both treatments.

Corrosion rate measurements following early test batches indicated that less flush was usually required than with conventional inhibitors. Only enough flush is required to carry the new formulation to the lower annulus walls of pumped-off wells. In wells with annular fluid levels, enough flush is required to carry the formulation down into the liquid column.

Fig. 2 (13345 bytes) depicts a typical annular inhibitor reservoir. The release mechanism appears to be both leaching from the metastable emulsion and shearing through the pump. Corrosion rate measurements and produced fluid samples confirm that the inhibitor is fully released at the surface.

Failure location in wells on conventional treatments compared with protection by the new treatments confirms that the inhibitor is released and available to inhibit corrosion throughout the entire well. For example, mud anchors and pumps also have improved protection.

ECONOMICS

Table 2 (11467 bytes) shows field-wide data from which economic information can be extracted. Failures include all rod and tubing-string failures plus pump changes. Most wells were rod pumped but a few had electric submersible pumps.

Economic evaluations assume all wells had rod pumps.

Failures were averaged for each field even though some wells typically gave more problems than others.

SHALLOW FIELD

To economically evaluate the decrease in pulling frequency, the first assumption is that a pulling job for relatively shallow wells averages $500 in direct cost. For example, the third field in Table 2 (11467 bytes) experienced a reduction to 1.16 from 1.34 pulls/well/year. For 134 wells, this equates to 24 fewer pulling jobs/year or a savings of $12,000.

If the oil price is $17.50/bbl, the average failure removes a well from production for 1.5 days. Therefore, this field should receive, after a 12.5% royalty, $1,820 more revenue/year, a significant amount in a stripper field.

With improved corrosion control, the tubing strings, rod strings, and pumps will last proportionally longer. In the field previously discussed, approximate costs are as follows:

$1,200-1,200 ft rod string
4,200-1,200 ft tubing string
$1,200 -pump.

For the pump and rods, expected service life had been about 1 year. Records now suggest that the life has doubled. Therefore, using a conservative 1.5 life increase yields an additional savings of $800/well/year.

In this field, the average expected tubing life is 5 years. Life expectancy should be extended at least by the pulling frequency improvement or $115/well/year.

Therefore, total operational savings for a year is $136,400 broken down as follows:

12,000-pulling reduction
$1,820-more oiling
$15,410-improved tubing life
107,200-improved rod and pump life.

For purposes of illustration, these wells may average $50/well/month in corrosion inhibitor treatment cost. With a 0.7 cost reduction factor, the savings would be $24,120.

Therefore, in this field, the total savings in operating and treatment costs is $160,500/year. This is significant savings in a stripper field where the corrosion inhibitor treatment cost might approach $70,000/year.

DEEPER FIELDS

Total savings vary from field-to-field. In deeper wells, string-life savings could be proportionally greater. ln more productive wells, value of added oil production would be proportionally greater.

The treatment interval in the first field listed in Table 2 (11467 bytes) was extended more than usual. This saved costs even though pulling frequency was not improved. In the Permian basin field (Table 2) (11467 bytes), failures declined to 0.81 from 0.95 failures/well/year. The most troublesome 24 wells in this field dropped to 1.20 from 1.69 failures/well/year.

For all 96 wells, the increased oil production would be worth $6,480. Pulling costs for this field, deeper than the shallow field example, are about $750/pull or a savings of $10,100/year.

In this field, the life expectancy of downhole equipment is not as well known as in the shallow field example. These wells are deeper but experience less pulling, therefore, the wells are assumed to be less corrosive.

For the shallow field, $122,600 was saved in improved rod, tubing, and pump life. If this figure is adjusted to account for fewer, deeper, less corrosive wells, the approximate savings for the Permian basin field would be $125,000/year.

Other annual savings are $6,480 for increased oil and $10,100 for reduced pulling cost.

If the average corrosion inhibitor treatment costs $50/well/month, savings are $17,300/year. Therefore, total savings of the new inhibitor program is $158,900.

OTHER CONSIDERATIONS

Although outside the scope of this article, other properties of the formulation also improve well performance and cost. Partly because treatments are less frequent, invoicing is simplified. The formulation also provides solids and scale control.

Compared to conventional inhibitor solutions, the formulation, because of its metastable emulsion nature, is also superior from an environmental standpoint.

ACKNOWLEDGMENT

The authors had considerable assistance in accumulating the information for this article, in particular, from Ed French and Bob Gray.

REFERENCES

Dvoracek, L.M, Corrosion, Vol. 32, No. 2, 1976, pp. 64-68.
Wu, Y. "A Binary Corrosion Inhibitor for Pumping Wells and Hot, Deep Gas Wells," NACE Conference, Calgary, March 1987.
Evans, S., World Oil, Vol. 210, No. 1, 1990, pp. 105-11.
French, E.C., Dougherty, J.A., and Martin, R.L. Reviews on Corrosion Inhibitor Science and Technology, NACE, 1993, pp II-1-1 to II-1-25.
Frank, W.J., OGJ, Vol. 74, No.23, May 31, 1976, pp 63-72.
McCoy, J.N., Paper No. SPE 9750, SPE Production Symposium, Oklahoma City, March 1981, pp. 93-108.
Annand, R.R., Corrosion, Vol. 22, No. 8, 1966, pp. 215- 28.
Martin, R.L., and Braga, T.G., Mat. Perf., Vol. 27, No. 2, 1987, pp. 16-22.
Martin, R.L., Mat. Perf., Vol. 22, No. 9, 1983, pp. 33-36.
Martin, R.L., Mat. Perf., Vol. 18, No. 3, 1979, pp. 41-50.
Rosenfeld, I.L., et al., Zashchita Metallov, Vol. 18, No. 2, 1982, pp. 163-68.