MEASUREMENT RECORDS VITAL FOR EFFECTIVE PROGRAM

July 9, 1990
Douglas P. Moore, Harry G. Byars ARCO Oil & Gas Plano, Tex. Corrosion monitoring is the foundation for a corrosion control program in oil and gas fields. The information derived is necessary to determine need, extent, and performance of corrosion control measures. This first part of a three-part series discusses the need to maintain good records on the corrosion control program and the techniques for measuring the corrosion environment. Emphasis will be on corrosion caused by produced
Douglas P. Moore, Harry G. Byars
ARCO Oil & Gas
Plano, Tex.

Corrosion monitoring is the foundation for a corrosion control program in oil and gas fields. The information derived is necessary to determine need, extent, and performance of corrosion control measures.

This first part of a three-part series discusses the need to maintain good records on the corrosion control program and the techniques for measuring the corrosion environment. Emphasis will be on corrosion caused by produced fluids.

SCORE KEEPING

Corrosion monitoring is the most important aspect of any corrosion control program. It is, in effect, score keeping. In each operating case, we decide how to address corrosion. The action may vary from no action to an involved integrated program using several control methods.

A score must be kept in each instance to determine if actions are prudent. Without monitoring, we are effectively blind and may not make correct decisions.

BASIC PRINCIPLES

Before discussing corrosion monitoring, it is important to understand some basic principles about corrosion. Corrosion is an electrochemical reaction resulting in deterioration of a component by its environment. In oil field applications, the necessary electrolyte is water. Water has to be present for corrosion to occur.

The variables influencing corrosion are complex and depend on the material exposed. In this series, we will address variables that influence weight-loss corrosion of metals, usually carbon and low-alloy steel. Weight-loss corrosion, in this case, is defined as the dissolution of metal ions into the electrolyte. This results in localized pitting or general mass loss.

Assuming water is present, the reaction rate of corrosion is dependent on temperature, pressure, carbon dioxide (CO2), hydrogen sulfide (H2S), pH, and oxygen (02). Other aspects of water composition such as chlorides, bicarbonate, and bacteria are important because of their influence on pH, oxidation, or conductivity. A more complete discussion of corrosion in oil field environments can be found in the literature.1

PROGRAM OBJECTIVES

Corrosion monitoring gathers information necessary to make decisions concerning corrosion control. Without monitoring, we cannot attack the corrosion problem efficiently. There are four basic reasons to monitor corrosion:

  1. To determine the need for corrosion control measures

  2. To evaluate the effectiveness of a corrosion control program

  3. To optimize a corrosion control program

  4. To detect changes in conditions and aid in troubleshooting problems.

The data obtained from corrosion monitoring will provide an adequate confidence level at the most economical cost. The scope of the monitoring program will depend heavily on the needs and economics of the system. This scope will vary not only from project to project, but change with time for a single project.

Multiple monitoring methods are advantageous. The benefits of getting a second opinion cannot be stressed enough. Monitoring methods vary in the type of data derived, quickness of response, and cost incurred. By using a blend of multiple methods, we can design a monitoring program to provide a firm basis for making decisions while maximizing the information-to-dollar ratio.

One key factor to successful monitoring is proper location and orientation of the monitoring device. Because there are many environments in a system, multiple monitoring locations may be required.

Many conditions can change in a system such as temperature, pressure, fluids separation, or flow rates. These changes can affect corrosion rates by altering such items as water content, velocity, CO2 solubility, H2S content, O2 entry, deposit formation, and microbial activity.

In every case, orient the monitoring device to expose it to the corroding medium.

RECORDS

Maintaining records is an important part of a corrosion monitoring program. These records provide information on past performance and present conditions. Computer systems are invaluable in this effort.2 3 4

Today's computer technology offers personal computers linked by networks with access to large mainframe computers. Without this help, data analysis can be laborious.

Even without computers, we must keep accurate records on producing conditions, corrosion control, and equipment performance. These data are a vital part of the monitoring program.

PRODUCING CONDITIONS

Records of producing conditions can aid in detecting changes in the system and, for example, differences between wells. Monitored producing parameters should be those with direct impact on corrosion rates. A few examples are flow rate, water cut, temperature, pressure, gas /water composition, and production method.

Because water is primary to corrosion, its amount and velocity has dramatic impact on corrosion rates. Changes in these conditions often require a shift in the corrosion control program.

Periodic gas analyses are needed to determine the level of components that affect corrosion in a system such as CO2, H2S, and O2, Changes could indicate upset conditions in a plant or intrusion of a foreign production zone. Early detection of these problems is necessary to reduce costly damage from corrosion.

Similarly, differences in water analysis can provide key information in anticipating, or showing cause for, a change in corrosion rates.

Variations in ion concentration (C1-, HCO3-, SO42-, etc.) can indicate intrusion of foreign water, changes in the source, or changes in the formation/condensed water ratio.

A periodic check of ion concentrations can indicate not only a change in corrosion rates, but may pinpoint the cause.

Other helpful water characteristics include pH, oxygen content, and bacteria presence. Maintaining records of these parameters is important because often it is not the value, but the change in value that is critical. The presence of bacteria, for example, does not show their activity. However, an increase in population count might.

The production method of a well will affect corrosion rates and monitoring choices. Conversion of a naturally flowing well to gas lift will increase the dissolved gases in the produced fluid. These gases can increase the corrosion rates substantially depending on the amount of CO2, H2S, and O2 in the lift gas. Therefore, records must be maintained of when the production method changed or will change.

EQUIPMENT PERFORMANCE

Information on installed equipment is equally important. Conclusions on whether a part has provided good service or failed, can only be drawn if the material type and conditions are known.

Equipment performance records are therefore an integral part of a corrosion control program. These records provide the definitive proof of program performance and apply to virtually any system.

When done properly, an equipment performance record system can equate to large savings in operating costs.2

An equipment performance monitoring program should have the following characteristics:

  • Commitment from management and field personnel.

  • A clear, concise form for recording the specific name of a system, well, etc.; service type; location within the system; failed component (maker, model No., etc.); cause of failure, if known; and repair method and cost.

  • A method for data assimilation and reporting.

The key to a successful equipment performance program is to provide a useful report back to those who supplied the data. If field personnel use the report, they will provide accurate, timely data for input. This combination of accuracy and usefulness will lead to continued support from field and management personnel.

One major step in achieving accurate records is proper design of the data form. The form must include all the information given above, but be short and easy to use. The easier to use, the more it will be used. The form should also be designed for easy transfer to a computer.

Obtaining the data is certainly a big part of the battle. Equally important, however, is the use of the data. Various types of reports may be generated to allow quick assimilation of trends. These may be as simple as a failure listing for a single well, or as complex as an annual company-wide statistical evaluation.

Using the data is best accomplished through a computer data base system that can extract information as tables, charts, or graphs. Computers allow easy assessment of the data, and in turn help make necessary changes in the corrosion control program.

Equipment performance records provide definitive data on the effectiveness of a corrosion control program. The program is easy and inexpensive to carry out. However, the time lag can be great, and waiting for failures can become very costly. For this reason, failure records are usually supplemented by other monitoring methods with quicker response.

CONTROL PROGRAM

Records must be maintained on the events and procedures of the corrosion control program. This information is used with other monitoring data to assess program effectiveness, Without it, optimization cannot be determined.

Data needed to optimize the program include:

  • When the program was implemented, or changed

  • What the actual program conditions and procedures were

  • If the program is effective.

For example, chemical treating programs may not be carried out as recommended. Maintaining proper treating frequency or amounts can be difficult even if the intent is genuine. However, the critical aspect is knowing the actual treating conditions. Only then can you make proper decisions based on other monitoring data.

MONITORING THE ENVIRONMENT

The following presents various methods for assessing the corrosive environment and monitoring changes.

WATER ANALYSIS

As discussed before, variations in the ion concentration of the water can indicate changes in the corrosiveness of the system. One way of tracking this is through periodic analyses of water samples. There are standardized methods which compare ion concentration graphically in a pattern, making trends more noticeable.-5 These standard analytical services are available at laboratories all over the world.

Sampling techniques are critical in getting representative water samples. For example, sampling ports should be flushed out to get a representative sample of a flowing stream. However, if you are concerned about corrosion in dead areas, the stagnant water in the sampling port is precisely what you need. Be sure the sample is representative and not contaminated by other factors.

In some cases, on-site analysis of certain ions is desirable. Unless kept anaerobically under pressure, reactions can occur in samples to change the equilibrium of some ions. Bicarbonate (HCO3-), for instance, converts to carbonate (CO32-) when dissolved CO2 leaves the solution.

Iron can oxidize to Fe2O3 unless the sample is preserved with acid. These changes can be critical. Onsite analysis is sometimes needed to avoid misleading results.

On-site analysis of various ions in water can be accomplished by using calorimetric kits or digital titration. The calorimetric kits produce a color showing presence and concentration. The results can be measured by using visual standards or a calibrated spectrophotometer.

Digital titrators are small, hand-held reagent dispensers. Various reagent cartridges are available depending on the ion in question. The digital titrator measures the amount of reagent needed to reach a visual end point.

The calorimetric tests can be performed on-site, quickly, and inexpensively. In most cases the spectrophotometer will provide more consistent results over visual comparison.

Digital titration is convenient and more reproducible. In some cases, these field methods may not be sufficiently accurate. Laboratory methods may be needed. However, you may need to preserve the field sample to get meaningful results.5

MEASUREMENT OF PH

Laboratory pH values for field water samples are not equal to the pH in the system. Because pH is a function of ions and dissolved gases, it can change drastically with time.6 Oxidation of iron followed by precipitation of ferric hydroxide can act to raise the pH. Loss of dissolved CO2 will also increase the pH. Therefore, pH must be measured on-site to be meaningful.

Field measurement of pH can be performed by indicating pH papers, or a pH electrode with meter. The papers are inexpensive and easy to use. indicating pH paper is available in various ranges.

The pH meters are more accurate, but can be subject to fouling. The pH meters can be costly; however, many new models suitable for field use are available at a reasonable cost. The accuracy needed depends on the system and its problems. In the oil field, an accuracy of 0.5 pH units is usually adequate.

ACID GAS ANALYSIS

Produced gas analyses are routinely performed in the laboratory by chromatography. A periodic analysis can indicate system changes that affect corrosion, such as CO2 content.

Analysis for H2S, however, must be conducted on-site. Metal sample containers will absorb H2S to varying degrees, resulting in lower values than actually in the field. Thus, a report of zero H2S in a lab sample has no significance.

Techniques for measuring H2S in the field include "length-of-stain" detector tubes,7 8 9 cadmium sulfate methods,10 11 and the Tutwiler method.12

The "length-of-stain" tubes are inexpensive and easy to use. However, the accuracy is operator dependent. The cadmium sulfate method is recommended for H2S concentrations less than 5 grains/100 scf (approximately 80 ppm). For greater concentrations, the Tutwiler method is preferred.

Often, the exact value of H2S is not as important as knowing if it is present. However, a precise value is required when determining the need for sulfide stress cracking (SSC) resistant materials. Based on accepted criteria, a system with an H2S partial pressure of 0.05 psi or greater requires SSC-resistant materials.13

Measurement of dissolved H2S in the produced water is also of importance. Detecting changes can pinpoint location of sulfate-reducing bacteria activity. It is helpful to be aware of dissolved H2S because it can interfere when conducting other analyses. Field test kits are commercially available that are easy to use and produce semiquantitative results.

OXYGEN

Oxygen presence is of significant importance in any system.14 It can enter through loose packing, ineffective seals, or open tanks. Dissolved oxygen in the water at levels greater than 0.025 ppm (25 ppb) can increase corrosion rates dramatically. Trace amount of oxygen in gas can create corrosion and safety concerns.

Dissolved oxygen content can be measured using a membrane probe oxygen meter or, more precisely, a membrane-covered polarographic oxygen detector (MPOD).15

A membrane probe is placed in the flow stream. Oxygen transports through the membrane due to the difference in partial pressure. The meter measures the current produced by the reduction of oxygen and correlates it to concentration.

There are many meters of this type on the market, some of which are capable of monitoring at levels and conditions pertinent to the oil field.

An MPOD can be used for continuous in-line monitoring in clean water systems. However, they require routine calibration and maintenance.

Calorimetric kits can also be used to measure dissolved oxygen in water. As with other calorimetric methods, results are dependent on field technique. Care should be taken to ensure no interfering ions are present. Errors in testing technique usually result in high readings. Therefore, calorimetries is often a handy go/no-go test.

A galvanic probe is one tool for detecting oxygen presence in water. The probe consists of two isolated, dissimilar metal electrodes, usually steel and brass. A sketch is shown in Fig. 1. In a water system, current will flow between the electrodes due to the potential difference between the dissimilar metals.

In time the electrodes will polarize, resulting in a decreased current. However, oxygen is a strong depolarizer. Current flow will remain high if oxygen is present, or increase anytime it enters.

The current output of the galvanic probe is not a quantitative measure of oxygen content. However, it is extremely useful in detecting cyclical entry of oxygen in a system.

Both the MPOD and galvanic probe are used for oxygen detection in water systems. They can work in oil/water systems as long as the sampling location avoids oil contact. Although oil can be detrimental to both methods, the MPOD is more susceptible to fouling than the galvanic probe.

Oxygen in gas lift and injection gas is also of importance to monitor. Oxygen present in these gases can accelerate corrosion in the gas system as well as the production. Oxygen in gas can be measured with a trace oxygen analyzer. Measuring trace oxygen at several locations can help pinpoint the cause of oxygen entry.

DEPOSIT ANALYSIS

Analysis of deposits found in a system can give needed information in addressing corrosion problems. Samples can be taken directly from piping or vessels, or from coupons exposed to the system. For example, solid samples can be caught when running a pig through a pipeline.

Knowing the compositions of these deposits can help determine the type of corrosion problem and detect changes in the system.

Sample collection and handling are important for proper interpretation of results. Select a representative sample and place it in a sealed container. The container must be labeled with the date, full details of the sample condition, and its location in the system.

Providing complete, accurate information is critical. Corrosion products can change after being removed from the system. For example, when iron sulfide comes in contact with air, it oxidizes to iron oxide. A sample that was black (iron sulfide) when collected, may be brown (ferric iron oxide) by the time it reaches the laboratory.

So, color of the sample when it was collected becomes important information. Sampling techniques designed to minimize oxygen contact are useful in avoiding these changes.

To determine composition, samples should always be analyzed in the laboratory. However, you can gain some immediate information with simple field tests. For example, place a small piece of the deposit in a cup and drop in a small amount of 15% hydrochloric acid (muriatic pool acid).

Record observations and include them in the sample description. If the sample reacts (fizzes) and gives off H2S (rotten egg odor), iron sulfide is present. If it reacts and no H2S is emitted, it is probably a carbonate. This information is very helpful to the laboratory in its analysis.

MICROBIAL ACTIVITY

Bacteria can increase corrosion problems by their presence or byproducts.16 17 The presence of bacteria colonies covering areas of the metal surface can accelerate corrosion by creating concentration cells. The bacteria can also create a local environment of low pH. When active, bacteria can change the environment and therefore influence corrosion.

The most common troublesome bacteria in oil field environments are sulfate reducing bacteria (SRB). SRB are anaerobic, that is they grow in the absence of oxygen.

SRB's can lay dormant in aerated solutions and become active after the oxygen is gone. They can also flourish in small oxygen-starved areas of an otherwise aerated system, like under deposits. SRB converts sulfates in the system to H2S, making the environment more corrosive.

There are many test methods to determine the presence and activity of SRB.5 18 19 The most common of these is a culturing method using API RP-38 broth medium.20

API RP-38 describes a serial-dilution method used to determine relative presence of SRB in water. The method gives a range of presence in colonies/ml, dependent on the number of broth bottles with a positive result.

Another method cultures SRB in a small amount of sand and nutrient.21

This procedure reduces environment disturbance by using a higher sample water-to-media ratio. An activity index is determined by watching the growth rate over a period of days.

Culturing from water samples will show presence of bacteria in the moving flow stream, i.e., planktonic bacteria. However, usually the sessile bacteria, deposited on the system parts, is responsible for corrosion.

Coupons or probes of various designs are used to study sessile bacteria activity.19 22 23 Their basic principle is to provide a surface where sessile bacteria can grow. For example, flush-mounted coupons can assess biocide effectiveness when running a pipeline pig. An analysis of the surface can reveal information on the type and activity of SRB.

Each type of bacteria monitoring has distinct advantages and disadvantages. Culturing water samples is easy but only captures the planktonic bacteria. In addition, false positive readings (within 2 hr) are possible in water containing dissolved H2S. Probes can provide information on the sessile population.

In either case, special strains of SRB may require a specific nutrient for detection. Above all, cleanliness is the key. Anything contacting the system must be sterile. Procedures are critical.

Presence of SRB does not constitute a problem. The important parameter is activity. Is the population growing, or is it stable? If SRB activity is found, look for a related problem. The presence of SRB is significant only if it causes a problem.

RESIDUAL CHEMICALS

Measurement of residual oil field chemicals can be helpful in trouble shooting a treating program. Sulfite residuals in the water can be used to help determine treatment dosages where sulfates or sulfur dioxide are used as an oxygen scavenger.

Chlorine residuals can be used to optimize chlorine treatment of fresh water for bacteria control. Field calorimetric kits are available for either of these applications. On-line monitors are also available for residual chlorine measurement. Detection of residual amounts of corrosion inhibitor is somewhat more difficult. There are some field and laboratory procedures used. Reliability is highly dependent on inhibitor chemistry and field fluids.

In many cases, laboratory techniques are the only choice. An increase in total amine can indicate an inhibitor is moving through the system. However, results are only qualitative.

A copper ion displacement test (CID) can help detect the presence of a filming inhibitor in a system.24 Here, a coupon is dipped in, or exposed to, the inhibited fluids and then immersed in a copper solution. Copper will deposit on those areas not filmed by the inhibitor. Examination can lead to a qualitative measure of inhibitor presence.

REFERENCES

  1. "Corrosion Control in Petroleum Production," TPC Publication 5, NACE, 1979.

  2. Bucaram, S.M., and Yeary, B.J., "A Data-Gathering System to Optimize Producing Operations: A 14-Year Review," paper No. 13248, SPE 59th Annual Technical Conference and Exhibition, Sept. 1619, 1984, Houston.

  3. Byars, H.G., and Gallop, B.R., "An Approach to the Reporting and Evaluation of Corrosion Coupon Exposure Results," Materials Performance, Vol. 14, No. 11, NACE, 1975, pp. 9-16.

  4. Galbraith, J.M., Hill, D.E., Bucaram, S.M., and Byars, H.G., "Utilizing a Computer-Based Corrosion Monitoring System at Prudhoe Bay, Alaska," paper No. 51. Corrosion/83, Anaheim, Calif.

  5. Ostroff, A.G., Introduction to Oilfield Water Technology, Chapter 2, NACE, 1979.

  6. Rittenhouse, G., Fulton, R.B. III, Grabowski, R.J., and Bernard, J.F, "Minor Elements in Oilfield Waters," Chemical Geology, Vol. 4, 1969, p. 189.

  7. "Standard Test Method for Hydrogen Sulfide in Natural Gas Using Length-of-Stain Detector Tubes," ASTM D4810-88.

  8. "Standard Practice for Determining Concentration of Hydrogen Sulfide by Direct Reading, Length of Stain, Visual Chemical Detectors," ASTM D4913-89.

  9. "Test for Hydrogen Sulfide and Carbon Dioxide in Natural Gas Using Length of Stain Tubes," GPA Std. 2377-86.

  10. "Standard Test Method for Hydrogen Sulfide and Mercaptan Sulfur in Natural Gas (Cadmium Sulfate Iodometric Titration Method)," ASTM D2385-81.

  11. "Determination of Hydrogen Sulfide and Mercaptan Sulfur in Natural Gas (Cadmium Sulfate Iodometric Titration Method)," GPA Std. 2265-68.

  12. "Hydrogen Sulfide in Gases by the Tutwiler Method," UOP Method 9-59, 1959, Universal Oil Products Co., Des Plaines, Ill.

  13. "Sulfide Stress Cracking Resistant Metallic Material for Oil Field Equipment," NACE MR-01-75.

  14. Byars H.G., and Gallop, B.R., "injection Water + Oxygen = Corrosion and Well Plugging Solids," paper No. 4253, Symposium on the Handling of Oilfield Waters, Dec. 4-5, 1972, Los Angeles.

  15. Hitchman, M.L., Measurement of Dissolved Oxygen, John Wiley and Sons Inc. and Orbisphere Corp, 1978, p. 59.

  16. NACE Basic Corrosion Course, 1970, pp. 1-11.

  17. Fontana, M.G., and Greene N.D., Corrosion Engineering, 2nd edition, McGraw-Hill Book Co., 1978.

  18. Tatnall, R.E.. Stanton, K.M., and Ebersole, R.C., "Methods of Testing for the Presence of Sulfate-Reducing Bacteria," paper No. 88, Corrosion/88, St. Louis.

  19. Patton, C.C., Applied Water Technology, Campbell Petroleum Series, 1986, Norman, Okla.

  20. "Recommended Practice for Biological Analysis of Subsurface Injection Waters," API RP-38.

  21. Bilhartz, H.L., "Bacteria Activity Index-A Practical Engineering Tool," S.W. District Division of Production API, Mar. 19, 1964.

  22. Ruseska, I., Robbins, J., Costerton, J.W., and Lashen, E.S. "Biocide testing against corrosion-causing oil-field bacteria helps control plugging," OGJ, Mar. 8, 1982, pp. 253-264.

  23. Galbraith, J.M., Mabile, N.J., and Van Buskirk, K.A., "Monitoring Sulfate Reducing Microbes at Prudhoe Bay, Alaska with Corrosion Coupons," paper No. 292, Corrosion/85, Boston.

  24. Hughes, W.B., "A Copper Ion Displacement Test for Screening Corrosion Inhibitors," Journal of Petroleum Technology, Vol. 10, No. 1, January 1958, p. 54

Copyright 1990 Oil & Gas Journal. All Rights Reserved.