Produced Water, And The Issues Associated With It
by Rodney Reynolds, Tertiary Oil Recovery Project, University of Kansas; Bob Kiker, PTTC Texas Permian Basin; and Lance Cole, PTTC National Project Manager
Excerpts in PTTC Network News, 3rd Quarter 2002

PTTC normally reserves this section for "state-of-the-art" articles developed by Karl Lang. Expect future contributions from Karl, but for this issue, PTTC is highlighting some practical field insights that are being documented as part of its DOE-supported PUMP (Preferred Upstream Management Practices) project in the South Midcontinent Region. When surveyed, operators in Oklahoma and Arkansas, not surprisingly, indicated that produced water and all the issues associated with it were primary factors constraining oil production. We're now developing a guidebook (to be completed later this year) and wanted to give you an "early peek" at what we're finding.

Tools for Identifying Water Entry

Produced water from each producer should be sampled and analyzed regularly. Knowing baseline trends, rapid, uncharacteristic changes become apparent. Chloride concentration reveals whether produced water is connate water, or water introduced to the well during stimulation or from other sources. Changes in chloride concentration may indicate poor mechanical integrity. Lower than normal chloride concentrations may indicate a shallow casing leak.

Production logs may be used for: (1) injection profile tests in waterfloods to determine the vertical distribution of fluid flows within the wellbore and the near wellbore region, (2) finding tubing-casing leaks,) (3) detecting lost circulation zones, (4) determining if packers or bridge plugs are leaking, (5) fluid channels behind casing, (6) production profiles, (7) gas-oil-water contacts, (8) to trace frac fluids, and are beneficial in many other instances.

Radioactive tracer logs help detect lost circulation, leaking packers and bridge plugs, fluid channels behind casing. They are frequently used to develop detailed injection and production profiles. Two types of radioactive tracer surveys are commonly used—the velocity shot method and the timed-run method. The velocity shot method is conducted by ejecting radioactive fluid downhole with a tool that has one or two gamma counter(s) and monitoring fluid movement with the gamma counters. A typical shot method procedure is to record one station above the perforations to check 100% flow and for any channeling above the perforations. The perforations are then surveyed in one to two foot increments until infinite time between counters is recorded and a second check made to ensure that no further "down channeling" is occurring.

The timed-run method qualitatively detects the flow of fluids up or down the hole, either in casing or in the annulus. In this method, a large amount of radioactive material is ejected at the bottom of the tubing and successive runs are made with a gamma-ray tool; the times of the ejection and each run are carefully noted. Movement of the radioactive material can be traced. This method is primarily used to detect any unwanted movement of the injected water in the casing annulus.

Differential temperature surveys measure the temperature of the wellbore fluid under either static (shut-in) or dynamic (flowing) conditions. The logging tool responds to temperature anomalies produced by fluid flow, either within the casing or in the casing annulus, and is very useful in detecting the latter. 

Interpretations are also used to determine flow rates and points of fluid entry or exit. In an injection well, temperature response is a function of depth, temperature of the injected fluid, injection rate, time of injection, formation and fluid thermal properties, and the geothermal profile. A well that has been injecting for some time can be shut-in, then repeat temperature logs run to observe the temperature profile as it returns towards geothermal values. The zones that have taken the (usually) cooler injection fluid will show a slower rate of return to the geothermal profile. This cooling of the geothermal profile can even detect uphole zones behind pipe that may be taking injection water due to communication problems.

Spinner (flowmeter) surveys are used to meter fluid flow rates within cased or uncased wells. They are useful in determining production rates, detecting thief zones, locating lost circulation zones, finding holes in casing or tubing, and assisting in injection and production profiles. The types of fluid flowing through a spinner can have a pronounced influence on their operation. Fluids that are dirty will foul the impeller movement, and gaseous fluids will over-spin the impeller. Higher viscosity fluids can cause readings to be high. Some spinners are limited to certain ranges of flow rates.

Water Shut-Off Treatment Using Gelled Polymers

For some producing wells, gelled polymer treatments are an option. In the U.S., most gelled polymer treatments in producing wells are in wells producing from fractured carbonate/dolomite formations that are associated with a natural water drive. Different polymer systems are available from different service providers. Recent successful treatments in the Midcontinent have used the MARCITsm technology developed by Marathon Oil Company. Service company experience seems to be a dominant factor in estimating how a particular formation in a given area will respond. Even then, service providers must be prepared to alter the original design based on the ability of a well/formation to accept a viscous fluid.

Pressure response during treatment provides key information. A slow, steady pressure increase over a period of time during pumping means one of two things - the problem zone is reaching polymer fill-up, or the reservoir temperature is causing the polymer to cross-link and build viscosity. Pressure response is a product of polymer volume, injection rate and gel strength. Altering any or all of these factors can improve the success of the treatment if reservoir resistance is not seen as the gelant is being pumped.

If the Hall plot indicates only a slight increase of pressure near the end of the treatment, service companies will typically recommend increasing polymer volume as a first step. Larger volume treatments lead to greater in-depth reservoir penetration, but they are more costly. Usually injection rates are increased at the beginning of the treatment to determine how easily the formation accepts a viscous fluid. Recent research and field experience has shown that higher pump rates can improve the effectiveness of treatments in carbonates that exhibit secondary permeability and porosity features. Increasing the injection rate also reduces the service company's field time, which lowers cost.

Increasing gel strength (viscosity) is another approach for achieving a pressure response. This is typically used at the midpoint of a treatment when the Hall plot shows no increase in slope. Gel strength can be increased by accelerating the cross-linking, increasing the polymer loading of the gelant, or using a higher molecular weight polyacrylamide. 

The best candidates for gelled polymer treatments are shut-in wells or wells producing at or near their economic limit. These types of wells benefit most from a successful treatment and little is at risk if the treatment fails, other than the treatment cost. Other selection criteria include significant remaining mobile oil in place, high water-oil ratio, high producing fluid level, high initial productivity, wells associated with active natural water drive, and high permeability contrast between oil- and water-saturated rock (i.e., vuggy and/or fractured reservoir). Successful treatments have been conducted in both cased and open hole completions.

Only empirical methods exist at this time for sizing treatments. Experience in a particular formation is important. In many instances, larger volume treatments appear to decrease water production for longer periods of time and recover more oil. Some rules of thumb include two times the well's daily production rate as the minimum polymer volume, or using the daily production capacity of the well at maximum drawdown (i.e., what the well would be capable of producing if it were pumped off) as the treatment volume. In lower fluid level wells, the daily production rate is sometimes used as the minimum polymer volume.

Before pumping treatments, ensure the wellbore is clean, acidizing if necessary. Establish a maximum treating pressure, run a step rate test to determine parting pressure. Select an acceptable source of water. Having the service provider test water compatibility is important. Select a polymer-compatible biocide (typically 5-10 gallons per 500 barrels of mix water). Set tubing and packer above the zone to be treated. While pumping the treatment, use stages of increasing polymer concentration. Inject the treatment at a rate similar to the normal producing rate. Keep treatment pressure below the reservoir parting/fracture pressure. Changing conditions during the treatment may warrant design changes during the pumping. Over displace the treatment with water or oil.

Wellbore Management (Reducing Failures)

If produced water volumes are high and one has to live with them, operating and maintenance practices that control and reduce costs are critical. Tubing failures are usually internal from either corrosion or rod wear, or external from buckling. Internal corrosion has to be controlled with the corrosion inhibitor program or an internal coating mechanism. There are some very basic practices to minimize tubing and rod wear. Some published literature indicates that tubing should always be anchored. However, many operators have found that tubing anchors are not needed in wells less than 3,000 ft deep. If the tubing failures are collar failures (external) from the tubing buckling, or if there is evidence of rod coupling wear on the inside of the tubing, then a tubing anchor is justified even at very shallow depths. Tubing anchors should be as close to the pump as practical. If the anchor is more than 400 feet above the pump, buckling can still occur.

Tubing rotators that distribute tubing are an option for shallower wells without tubing anchors, but they are used infrequently. A new, lower cost mechanical rotator developed by Omega Technologies is being used in shallow wells (2,000 ft) with reported good results. Rod rotators can be used to distribute coupling wear. They are only used when rod guides are used. If a rod rotator is to be used, the operator should be sure that the correct rotator is used. 

Since increased fluid velocity around rod guides can remove inhibitor films, rod guides should only be used where repeated tubing splits and/or excessive rod coupling wear occur. Often, wear is concentrated on the bottom of the rod string where rods go into compression. An accepted practice is to install 3-4 guides per rod with the number varying with hole deviation. When tubing wear is found at the bottom in the sinker bar area, a short, 4 ft guided stabilizer is run between each sinker bar. Operators prefer "molded-on rod guides" over "snap-on rod guides," since snap-on rod guides require hammering them on, which can damage rods.

Polyethylene tubing liners are becoming increasingly popular. These liners are chemically inert and are a seamless tube tolerant to minor surface imperfections. Some operators are installing the liners one or two joints above the pump, while others are installing the liner in half of the tubing string. Polyethylene tubing liners can be installed in used "green and or blue band" tubing for a cost of about $1.50 per foot. There is some ID loss, but favorable friction characteristics partially offset ID losses.

Sucker Rod Handling

There are many, many variables that can lead to sucker rod pin and coupling failures. These are under torque, wrench marks, over torque, lubrication, contamination, thread wear and cross-thread. Be prudent when picking up rods. Thread protectors should be unscrewed, because knocking off the thread protectors leaves plastic remnants that can cause damage during makeup. When tailing rods to the floor, take care to prevent metal-to-metal contact when rods are drug, or contaminating the threads by dragging ends through dirt. Inspect and clean, as needed, all rod pins and boxes. Once the pins and boxes are clean, use only moderate lubrication.

All rod connections should be made "hand tight" using "only hands;" not hands holding wrenches! If a rod will not screw on by hand, thread damage has already occurred. Once hand tight, a mark can be placed across the connection to represent the first point of a distance of travel. This line represents zero displacement. Then, before tightening with tongs, the rod should be backed off by unscrewing approximately four rounds to allow the tongs time to reach full speed, ensuring the momentum force component of makeup is comparable to normal operating condition of the tongs.

Reducing Well (Rod/Tubing/Pump) Failures

Operating practices to reduce well failures vary from "none" (i.e., replace a single broken rod) to "very disciplined well failure analyses and corrective action processes." Field experience confirms that more thorough and disciplined programs reap results, as illustrated for one West Texas field. 

Well failure frequency averages for successful programs are generally under one failure per well per year, and some are as low as 0.15 failures per well per year. One might argue—that's great, but I'm a small company. The good news is that smaller companies can apply the same concepts by employing the services of their vendors and maintaining good communication.

Common elements of successful programs are:

• Visual inspection of rod and tubing failures
• Discussion of the problem and a review of the well history with team members-company, chemical, service company, equipment vendors, rod and tubing inspection personnel.
• Establishment and maintenance of a database of each well's failures. A tracking program can set up the database. The tracking program should also facilitate economic evaluation.

Over time, best operating practices in a given area should be developed. Successful programs usually employ the "pay me now, or pay me later" philosophy. Example; if a rod fails, successful programs call for either replacing the entire rod string with an inspected used rod string, or replacing the tapered section where the failure occurred. If one rod failed for mechanical or corrosive reasons, chances are that another failure will occur shortly. The pulled rod string is sent in for inspection. Typical inspection costs are: $7.95 per good rod, and $2.00 per rod rejected. A similar practice applies to tubing string failures. Used tubing strings of Yellow, Blue, and Green pipe body ratings are reused after inspection. Polyethylene liners work well in used tubing strings.

Lift Efficiency and Type of Lift System

With any lifting system, "system efficiency" is very important. Overall system efficiency is defined as the amount of theoretical work required to lift the liquid from the net liquid level depth to the surface divided by the amount of power supplied to the motor. There are programs that evaluate system efficiencies. One system offered by Echometer is computerized and takes about 45 minutes per well. System efficiency quickly translates into power cost savings.

For any lifting system, experience indicates that failure frequency is the most important variable. The other components of the system—equipment, servicing, power consumption cost—are much more stable, especially on beam-lifted wells. When considering typical failure frequencies, power consumption, service cost and equipment costs, general conclusions can be drawn. For producing depths of 4800 ft and wells with 5 ½-in casing considering "Total System Cost", beam pumps are most economical for volumes up to 320 BFPD. However, for this same depth and casing size, when considering only the "Operating and Maintenance Cost", the volumes where beam pumps are most economical increase to about 500 BFPD. For volumes above 500 BFPD, electrical submersible pumps (ESPs) are generally most applicable. ESPs are also preferred when holes are extremely crooked.

Kent Gantz, "Holistic Producing-Well Improvement
Reduces Failures/Servicing Costs, "Fig. 1, Petroleum
Technology Digest section of World Oil, June 2002, p. 59-60. 

J.N. McCoy et.al, "Modern Total Well Management,"
Figure 1 from SPE #62834 presented at 2000/SPE/ AAPG
Western Regional Meeting, Long Beach, CA, June 19-23, 2000