Review of Biotreatment, Water Recovery, and Brine Reduction Systems for the Pueblo Chemical Agent Destruction Pilot Plant (2013)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

4 Materials of Construction Review MATERIALS OF CONSTRUCTION FOR THE Finding 4-2. The combination of high chloride concentra- BIOTREATMENT, WATER RECOVERY, AND BRINE tions and elevated temperatures creates highly corrosive REDUCTION SYSTEMS conditions, even when the pH is above 7. Overview The diversity of materials may present issues when repairs are required because the methods for repairs and The biotreatment, water recovery, and brine reduction replacements differ for various MOCs. For example, weld- systems (WRS-BRS) include a large number of different ing methods and filler metals for austenitic stainless steels items that are constructed from a wide spectrum of materials, (the 300-series stainless steels) are very different from those including carbon steel, high nickel alloys, fiber-reinforced used for duplex stainless steels. Nickel alloys require yet plastic (FRP), epoxy coated carbon steel and titanium, and different methods. austenitic and duplex stainless steels, the latter two being used extensively. A list the materials of construction (MOCs) Finding 4-3. The methods for repairs and replacements dif- for the major pieces of equipment used in the WRS-BRS, fer for various materials of construction. Therefore, the diver- along with MOCs used in some other PCAPP equipment, sity of materials used in the PCAPP biotreatment system and is given in Appendix A. The vendor’s MOC listing for the WRS-BRS may present issues pertaining to material logistics biotreatment system is given in Table 4-1. and appropriate craftsmen skills when repairs are required. The committee did not identify any obviously inappro- priate MOC selections, in the limit of the information that Heat transfer surfaces are especially vulnerable to attack. was provided other than that the biotreatment steam supply Furthermore, the possibility of forming deposits on some sur- p ­ iping is coated with a paint system rated for 120°F maxi- faces was noted.1 These deposits may be precipitated solids mum (see footnote to Table 4-1). This piping will operate from the process fluids or corrosion products from upstream above this temperature. equipment. Such deposits act as a shield that creates a stagnant condition under the deposit. They also insulate the Finding 4-1. The materials of construction selected for the surface and can lead to higher metal temperatures. The dis- PCAPP biotreatment and WRS-BRS process equipment solution of metal in the crevice under the deposit causes an appear to be generally appropriate for the service condi- increase in positively charged metal ions. This charge is bal- tions anticipated, based on the information available to the anced by the migration of chloride ions into the crevice. The committee. metal chloride that forms then hydrolyzes to form insoluble metal hydroxides plus HCl, thereby lowering the pH in the The committee cannot rule out corrosion problems, par- crevice (pH ~2 to 4). This promotes further metal dissolution, ticularly if transient events lead to extreme off-normal condi- which results in a rapidly accelerating autocatalytic process tions. Large portions of the equipment surfaces are exposed of metal attack. PCAPP staff have emphasized the desirabil- to high chloride ion concentrations. Combined with elevated ity of frequent cleaning to remove deposits.2 temperatures, this presents highly corrosive conditions, even though the liquid stream is above pH 7 during much of the 1George Lecakes, Chief Scientist, PCAPP, “PCAPP’s Water Recovery process. Stress corrosion cracking may also occur, even System and Brine Reduction System Briefing,” presentation to the com- though many of the materials are resistant to, but not immune mittee, May 1, 2012. from, this mode of failure. 2Ibid. 33 

34 REVIEW OF BIOTREATMENT, WATER RECOVERY, AND BRINE REDUCTION SYSTEMS FOR PCAPP TABLE 4-1  PCAPP Biotreatment Area Materials of Construction Area of System Material of Construction Coating Biological Treatment System (BTS) BTS air feed piping Carbon steel Exterior painta 25% caustic piping Carbon steel Exterior painta Steam supply piping Carbon steel Exterior painta Immobilized cell bioreactor (ICB) tank Carbon steel Interiora/exterior painta Feed tank Carbon steel Interiora/exterior painta Effluent tank Carbon steel Interiora/exterior painta Feed tank agitator Carbon steel with 2205 stainless mixer, shaft, Motor/mounting painted (vendor standard) and blade Feed tank catwalk grating Carbon steel Galvanized ICB catwalk grating and stairs Carbon steel Galvanized Equipment skids Carbon steel Exterior painta Pump materials CD4MCuN Exterior paint (vendor standard) Moisture separator 304 Stainless steel None Undiluted hydrolysate piping Carbon steel Exterior painta Diluted hydrolysate piping PVC Schedule 80b None Biologically treated hydrolysate piping PVC Schedule 80b None Process water piping PVC Schedule 80b None Reverse osmosis reject piping CPVC Schedule 80b None Nitrogen nutrient piping CPVC Schedule 80b None Phosphate nutrient piping CPVC Schedule 80b None Off-gas treatment system piping prior to CPVC Schedule 80 b/carbon steel None moisture separator Off-Gas Treatment System (OTS) Carbon filter (granulated activated carbon) Carbon steel Exterior and interior epoxy urethane (vendor standard) OTS piping after moisture separator Carbon steel Exterior painta OTS fan SAE 960X steel wheel Exterior paint (vendor standard) OTS fan exhaust stack Carbon steel Exterior painta a Sherwin-Williams Phenicon® HS Epoxy coating system to be applied to interior and exterior of equipment per Golder Cleaning and Coating Procedures (2485-V1A- MS00-0044s01). b Polyvinyl chloride (PVC) and chlorinated PVC (CPVC) shall be insulated for ultraviolet protection. SOURCE: PCAPP responses to committee questions submitted January 16, 2013, dated January 24, 2013; specification table originates from Golder Associates. Finding 4-4. Heat transfer surfaces are vulnerable to cor- the 4- to 5-year span during which the plant is intended to rosion, especially under deposits. The need for surfaces of operate. Thus, the committee believes that an appropriate the PCAPP biotreatment system and the WRS-BRS to be course of action for the PCAPP operators is to focus on cre- frequently cleaned to remove deposits has been recognized ating rigorous corrosion-monitoring (testing and inspection) by PCAPP staff. and maintenance protocols for these units during systemiza- tion and use these protocols during operations. Recommendation 4-1. Standard operating procedures should be implemented to ensure that appropriate surface Finding 4-5. Properly designed and implemented protocols cleaning is performed regularly for the PCAPP biotreatment to monitor the occurrence of corrosion in components of system and the WRS-BRS. the PCAPP biotreatment system and WRS-BRS enables the identification of problems before equipment failures occur. This provides a basis to select alternative materials if the Corrosion Monitoring original selections prove to be inadequate or modify operat- The biotreatment, WRS, and BRS equipment is already ing practices to minimize the corrosion risk. installed, and, as previously mentioned, the committee has not identified any inappropriate materials selections. The PCAPP has developed a corrosion-monitoring plan for materials in these units may well perform suitably if such tank systems regulated under the Resource Conservation considerations as operational stresses, fluid velocities, and and Recovery Act that uses visual inspection of the exte- area ratios of galvanic couples are properly managed. How- rior surfaces, document reviews, ultrasonic nondestructive ever, the composition of the liquid stream is unique and leads inspections, and calculation of corrosion rates and remaining to uncertainties in the rate at which corrosion can occur over life (BPT, 2012). However, it does not incorporate any cor- 

MATERIALS OF CONSTRUCTION REVIEW 35 rosion testing methods, nor does it include visual inspection that PCAPP develop a corrosion-monitoring plan for equip- of the interior surfaces. Titanium equipment and piping are ment other than tanks that includes, as a minimum, internal excluded from monitoring on the basis that this material is inspection and corrosion testing of all equipment, including unlikely to corrode during the time that the system is in use. that made of titanium. This plan should include provisions Also, heat exchangers are not included in this plan, and there for detecting general corrosion, pitting/crevice corrosion, are no plans to develop a plan for this equipment.3 and stress corrosion cracking. The committee has concerns with the proposed corrosion- monitoring plan. Ultrasonic nondestructive testing can Recommendation 4-3. A corrosion-monitoring program detect changes in wall thickness from general corrosion or for the PCAPP ICBs and WRS-BRS should include the cracking, but it is not reliable for detecting pitting, which following: may be the most likely form of corrosion in this system. Reliance on exterior visual inspection will not warn of dam- • Corrosion monitoring conducted on an ongoing basis age until the equipment wall has been penetrated. Internal to identify corrosion problems and to provide data visual inspection of the equipment when it is out of service for selecting alternative materials. This may require between destruction campaigns4 could reduce the probability using non-standard monitoring methods for the heat of unexpected failure. Use of one or more of the corrosion- transfer surfaces. monitoring techniques described below would further reduce • Nondestructive testing of the equipment and piping the chance of unexpected failure. on a scheduled basis, with special attention given to The absence of a corrosion-monitoring plan for heat heat transfer surfaces. exchangers is an even greater concern. The heat transfer surfaces are thin and operate at the highest temperatures. The creation and implementation of a comprehensive The probability of fouling has already been identified. These monitoring program will require the full support from PCAPP heat transfer surfaces are vulnerable to pitting and crevice management. This would involve making provisions for the corrosion attack under deposits. Even a small pinhole leak added time and expense incurred in performing corrosion in one of these heat exchangers would shut down the system. monitoring to be accounted for in the rewards system under As a minimum, the heat transfer surfaces should be inspected which the site operations contractor works. Unless such top- by a boroscope or comparable device whenever possible. level buy-in occurs, it is unlikely that a corrosion-monitoring Titanium surfaces should be included in the corrosion- program will be successfully implemented because many monitoring plan. distractions may take priority during day-to-day operations in a complex plant such as PCAPP. The consequences of Finding 4-6. The proposed tank corrosion-monitoring plan failing to implement a corrosion monitoring program could for PCAPP relies on ultrasonic inspection and external be serious, including the possibility of structural failures, visual inspection. This approach would provide no warning with consequent impacts on safety and the environment, and of pitting attack prior to wall penetration. Internal visual of operational failures with impacts on cost and schedule. inspection of the equipment when it is out of service between Agarwala and Ahmad (2000) report that “a major portion of destruction campaigns would reduce the probability of such the lifecycle cost for all platforms and infrastructures is due failures. to labor hours spent in finding the problems and then fixing them with either a major overhaul or extensive part replace- Finding 4-7. No corrosion-monitoring plan has been devel- ment” (p. 1). They found that a monitoring program could oped for equipment other than for tanks. Equipment items save up to 30 percent of the total cost. other than tanks, especially heat exchangers, are vulnerable to attack—to pitting and crevice corrosion attack in par- Finding 4-8. Historic evidence finds that the use of a moni- ticular. Even a small leak could shut down a system having toring program saves up to 30 percent of the total costs. such equipment and, potentially, the entire PCAPP facility. A corrosion-monitoring plan for such equipment is required. Recommendation 4-4. The Assembled Chemical Weapons Alternatives program leadership and PCAPP site contractor Recommendation 4-2. An inspection plan for the internal management should strongly support implementation of a surfaces of tanks beyond the use of ultrasonic monitoring comprehensive corrosion-monitoring program. This should should be developed. The committee strongly recommends be a priority. 3Teleconference between George Lecakes, Chief Scientist, PCAPP, and CORROSION-MONITORING METHODS the committee on September 17, 2012. 4There typically are scheduled downtimes during which the equipment As an element of an enterprise’s asset management is recalibrated and reconfigured to process a different type of munition. A activities, an important role of corrosion monitoring is to campaign refers to operations relating to a particular type of munition and/ or a particular type of agent. focus efforts toward preventive and predictive maintenance, 

36 REVIEW OF BIOTREATMENT, WATER RECOVERY, AND BRINE REDUCTION SYSTEMS FOR PCAPP and away from more costly corrective maintenance prac- be hard-wired to a control room system. In the latter case, tices (Tullmin and Roberge, 2000). The following sections readings can be taken automatically and the results converted describe common methods of monitoring corrosion in equip- to corrosion rates via software. ment MOCs. There are some disadvantages to using electrical resis- tance probes. Vessel or piping walls must be penetrated to install the probes, and this provides an opportunity for Use of Metallic Coupons creating leaks. Short time duration between readings can Using metallic coupons in a rack is the simplest and most result in erroneous rate calculation if the corrosion rate is common corrosion testing method used in plant facilities. low. Additionally, the method provides no information on These coupons are placed in a rack at selected locations in the localized corrosion, which is the most likely mode of attack processing equipment throughout the plant where they are in this process. exposed to the feed stream and are periodically examined to measure the corrosion. These coupons typically include the Polarization Resistance Measurement materials used in the process equipment and other ­ aterials m with greater corrosion resistance that might be selected if the Polarization resistance measurement methods provide an chosen materials prove to be inadequate. Frequently, dupli- estimate of the corrosion rate. Polarization resistance is an cate coupons are included to provide greater confidence in electrochemical (current) response to an imposed driving the results. This test has the advantage that many materials force (an electrochemical potential). This method is based on can be tested at the same time, and it is fairly inexpensive. the Stern-Geary equation (Stern and Roth, 1957). The theory Crevice corrosion tendencies can be detected by using behind this technique is based on the corrosion rate of a probe specially designed washers on the coupons. Welded samples being inversely proportional to its polarization resistance, can be included to determine if weld or weld heat-affected that is, the slope of the potential-current response curve near zone corrosion is a problem, and stressed samples can be the steady state corrosion potential as shown in Figure 4-1. used to detect stress corrosion cracking. As the slope increases, the corrosion rate decreases. Disadvantages to using coupons include not detecting The probe electrodes are fabricated of the material being changes in corrosiveness with time and not detecting heat tested. An electronic power supply polarizes the specimen transfer effects. Also, the plant must be shut down to gain about 10 mV from the corrosion potential. The resulting access to the coupons for examination. This disadvantage current is recorded as a measurement of the corrosion rate. can be overcome by using retractable coupon holders. NACE This method yields an instantaneous estimate of the corro- International and ASTM International have issued standard sion rate (Dean and Sprowls, 1987). A typical three-electrode practices for conducting these tests. See Appendix B for a polarization resistance probe is shown in Figure 4-2. listing of the NACE and ASTM standard practices and guides Polarization resistance measurement has some of the and other sources of materials. same disadvantages as electrical resistance probes: (1) the wall of the equipment must be penetrated, and (2) this method does not indicate localized corrosion. Electrical Resistance Probes Some practitioners have found that changes in corro- Electrical resistance probes may also be used to measure sion potential are a useful indicator of pitting tendencies.5 corrosion in process equipment. These probes are specially Increases in the corrosion potential toward the transpassive designed corrosion coupons. The corrosion rate is calculated region are an indication of possible breakdown of the passive from the change in electrical resistance rather than from mass layer on stainless steels or nickel alloys. Measuring the corro- loss. These measurements are made by installing a wire or sion potential requires the use of a stable reference electrode tube fabricated of the material in question in such a way in the process stream, and such electrodes are available. that the electrical resistance can be conveniently measured. Corrosion reduces the cross section of the probe; therefore, Electrochemical Noise its electrical resistance will increase with time if corrosion is taking place. A temperature-compensating element should The electrochemical noise methodology for measuring be incorporated into the probe, since changes in temperature corrosion involves the measurement of electrochemical noise also influence the resistance. Electrical resistance probes produced by the corrosion process. This is accomplished by measure the average remaining metal thickness. the following: To obtain a corrosion rate, a series of measurements are made over time, and the results are plotted as a function of • Measuring potential noise by monitoring the poten- exposure time. The corrosion rate can be determined from tial difference between two electrodes, or the slope of this plot. Commercial electrical resistance equip- ment is readily available from a number of vendors. Measure- 5Personalcommunication between S.W. Dean, Engineer (retired), and ments can be made with portable devices or the probes can committee member Robert Puyear, via telephone, on May 14, 2012. 

MATERIALS OF CONSTRUCTION REVIEW 37 FIGURE 4-1  A typical electrical resistance probe. SOURCE: Reprinted with permission of the American Water Works Association from the 1995 Water Quality Technology Conference; permission conveyed through Copyright Clearance Center, Inc. FIGURE 4-2  Three-electrode polarization resistance probe (a) and schematics of installation of the probe in pipe fitting (b), in a welded line (c), and in a pipe tee (d). SOURCE: Dean and Sprowls (1987). Reprinted with permission of ASM International. All rights reserved. www.asminternational.org. 

38 REVIEW OF BIOTREATMENT, WATER RECOVERY, AND BRINE REDUCTION SYSTEMS FOR PCAPP • Measuring current noise by monitoring current fluc- should be taken before the equipment is put into service, and tuations of two identical shorted electrodes by using the location of the readings should be clearly marked on the a zero resistance ammeter, or equipment. Subsequent readings can be taken at these same • Using both of the above methods. points over time to detect metal loss. Whatever inspection techniques are used, it is critical that The potential and/or current signals are compared to complete records are made and maintained. These records, provide information on the corrosion process (type of cor- along with the results of the corrosion-monitoring program, rosion, time of corrosion initiation, and, in some instances, provide information critical to decisions about equipment corrosion rate). Electrochemical noise was originally repair or replacement. designed to monitor localized corrosion, such as pitting By itself, corrosion monitoring data has a relatively low and crevice corrosion (Agarwala and Ahmad, 2000). ASTM utility. Its value comes when it is incorporated into the plant’s has developed a guide for conducting electrochemical information system and integrated with inspection data, noise tests (ASTM G199-09, 2009). This guide covers operational parameters, failure statistics and analysis reports, the procedure for conducting online corrosion monitoring maintenance records, and process chemistry data. of metals by this technique, which can be used to detect localized corrosion activity and to estimate corrosion rate REFERENCES on a continuous basis without removal of the monitoring probes. The ASTM guide also provides some generally Agarwala, V.S., P.L. Reed, and S. Ahmad. 2000. Corrosion Detection and Monitoring—A Review. Houston, Tex.: NACE International. accepted methods of analysis that are useful in interpreting ASTM (American Society for Testing and Materials). 2009. ASTM G test results. A detailed discussion of electrochemical noise 199-09 Standard Guide for Electrochemical Noise Measurement. West measurement and signal analysis is beyond the scope of this Conshohocken, Pa.: American Society for Testing and Materials. report, but some of the basic principles are discussed by BPT (Bechtel Pueblo Team). 2012.  Corrosion Monitoring Plan for RCRA Roberge and Klassen (2000). Examples of the application Tank Systems. 24852-RD-30G-000-V0001, Rev. 00C. Pueblo, Colo.: Bechtel. of electrochemical noise measurements in the solution of Dean, S.W., and D.O. Sprowls. 1987. In-Service Monitoring. P. 200 in engineering problems are presented in papers by Lawson et Metals Handbook, Volume 13: Corrosion; Corrosion Testing and Evalu- al. (2000), Wharton et al. (2000), and Goeller et al. (2000). ation. Metals Park, Ohio: ASM International. Goellner, J., A. Burkert, A. Heyn, J. Hickling, and H.U. Volger. 2000. Using Electrochemical Noise to Obtain More Information from Conventional Other Considerations Relevant to Corrosion Monitoring Corrosion Test Methods. Houston, Tex.: NACE International. Lawson, K., G.E.C. Bell, and G. L. Edgemon. 2000. Integrity Management A number of other testing methods are available. Bypass Utilizing Electrochemical Noise—Practical Experience. Houston, Tex.: loops can be used to test corrosion on heat transfer surfaces. NACE International. The bypass stream is passed through a small heat exchanger Roberge, P.R., and R.D. Klassen. 2000. Electrochemical Noise Analysis for where the tubes are the test samples. Visual inspection of Corrosivity Assessment. Houston, Tex.: NACE International. Stern, M., and R.M. Roth. 1957. Anodic behavior of iron in acid solutions. both the exterior and internal surfaces should be conducted Journal of the Electrochemical Society 104(6):390-392. on a planned schedule and whenever an opportunity is Tullmin, M.A., and P.R. Roberge. 2000. Monitoring Corrosion in Aging presented, such as an unplanned or routine maintenance Systems—New Possibilities and Old Fundamentals. Houston, Tex.: shutdown. Boroscopes are very useful for examining the NACE International. interior of heat exchanger tubes and critical piping elements. Wharton, J.A., B.G. Mellor, R.J.K. Wood, and C.J.E. Smith. 2000. Crevice Corrosion Studies Using Electrochemical Potential Noise Measure- Ultrasonic thickness measurement is a useful tool for detect- ments. Houston, Tex.: NACE International. ing changes in wall thickness. Base-line thickness readings