It is essential to understand how crud build-up leads to elevated cladding temperatures and abnormal cladding corrosion for Boiling Water Reactor (BWR) fuel. With a heated rod test loop configuration, the influence of water chemistry on crud build-up and cladding corrosion behavior can be assessed. A new test loop named the Westinghouse Annular Thermal Crud Hydraulic (WATCH) Loop, has been designed to grow simulated BWR crud under normal BWR operating conditions. WATCH simulates the bottom meter of a BWR core in terms of temperature, flow, steam quality and heat flux. Test results from this loop can be used to benchmark risk assessment tools, which in turn are used to make predictions and minimize or avoid BWR fuel rod failures, consistent with guidelines from the Institute of Nuclear Power Operations (INPO) and the Electric Power Research Institute (EPRI). In this paper, the Westinghouse Annular Thermal Crud Hydraulic (WATCH) Loop design and development are introduced. The results of a commissioning test and a run producing zinc-rich deposits are also presented.

Water chemistry plays an important part in maintaining corrosion resistance in water transport systems throughout nuclear power plants (NPP’s). Small changes in liquid chemistry such as pH, borate concentration, or build-up of crud in reactor cooling water can result in rapid degradation or damage to components and lead to unexpected failures. The Chemical and Volume Control System (CVCS) and Reactor Water Cleanup System (RWCU) are responsible for maintaining these parameters at appropriate levels, and so failure of either of these systems can result in unnecessary stresses on many other reactor systems due to resulting transients. While the major components of these systems all have sufficient redundancy to prevent major accidents, failure of components in these systems can result in failure of other redundant components and affect plant safety [1]. The CVCS and RWCU systems have experienced aging related degradations and failures in the past, and although they have not affected the system’s emergency functions, they have resulted in unnecessary actuation of related systems, and reactor shutdowns [1]. Reactor shutdowns can result in large changes in reactor coolant chemistry such as oxygen and borate concentration transients, and the build-up of corrosion products which can’t be as easily removed during periods of reactor shutdown [2]. In the following analysis of Component Operational Experience Degradation and Ageing Program (CODAP) experience data; causes, impacts, and preventative actions as recorded in CODAP are examined for degradation events which took place in the CVCS and RWCU, of PWRs and BWRs, respectively. The analysis will demonstrate the usefulness of CODAP in examining reactor component failure trends, as well as discuss insights on improvement for the program.

Two Zr-Sn-Nb alloys with minor Germanium or silicon additions were prepared by traditional manufacturing process to meet the design requirements. Transmission electron microscope (TEM) and scanning electron microscope (SEM) were utilized to characterize the detail microstructure of base alloys. Corrosion resistance was examined by the weight gain in static autoclave with different water chemistry environments. The mechanical properties at room temperature and elevated temperature were evaluated by conventional tensile testing. Thermal creep resistance was evaluated by an internally-pressurized creep test at 385 °C with hoop stresses of about 108 MPa and 150 MPa (during 24 h). It was found that SZA-6 and SZA-4 alloys consisted of partially recrystallized grain structures with uniformly distributed fine second phase particles (SPPs) located within grain interior and at grain boundaries. Both SZA-4 and SZA-6 alloys exhibited excellent corrosion resistance in two water chemistry conditions. The corrosion resistance of SZA-6 was better than the reference commercial alloy, and SZA-4 was slightly better than SZA-6. The mechanical properties of two new zirconium alloys were comparable, and both of them can meet the design criterion. Moreover, the thermal creep resistance of SZA-4 and SZA-6 alloys was equivalent to existing commercial alloy. Considering the outstanding corrosion resistance, satisfied mechanical properties and thermal creep resistance, SZA-4 and SZA-6 alloys were suggested as promising alloys used for CAP1400 fuel assembly in the future.

In chemical researchers view, hot functional test is a verification of the Nuclear Power Plant before first fuel loading and commercial operation, which is the preparation for staffs, documents, instruments and sampling systems. So, chemistry department should use its own language, knowledge and experience to express their thoughts and what they have seen during the engineering commissioning period. As the first commercial operation nuclear power plant after Fukushima nuclear accident, during the four units commissioning period, chemical researchers accumulated a lot of good experience and feedbacks in the aspect of construction and commissioning for new nuclear power plant. For example, in order to ensure the personnel skill level, we must make special plans which include staff training, laboratory construction, instrument and on-line system commissioning, and all of these should be timely adjusted and changed in order to be consistent with the engineering progress. In order to ensure the water quality of pipe flushing in different stages, such as cold functional test, hot functional test, we should set a strictly water chemical standard which based on the HAF103, and the standard should have some differences in different stage for one unit. In order to ensure the water chemistry in good performance especially after the unit going into commercial operation, the maintenance plan for equipment and system must be formulated, and then, a detailed monitoring plan must be executed. At the same time, a strict system flushing controlling mode can also provide a great benefits for water chemistry quality, especially in the period of start-up. In addition to these above experiences, chemistry researchers of Ningde nuclear power plant also accumulated a lot of good practices and feedbacks about dealing with some abnormal water quality activities, which can’t be founded in commercial operation unit. For example, the aluminum (Al) content in the primary increased rapidly and beyond the specification limits in hot functional test and so on. This article will share the good practices and feedbacks of the first phase of Fujian Ningde nuclear power plant. We hope these good practices and experience feedbacks can provide good reference for the other new nuclear plants in the stage of design, construction, operation and maintenance in the future.

Generic investigations regarding the influence of corrosion processes of hot-dip galvanized PWR containment installations on strainer clogging as well as on the coolant chemistry and possible resulting in-core effects are carried out within joint research projects of the Helmholtz-Zentrum Dresden-Rossendorf (HZDR), TU Dresden (TUD) and Zittau-Görlitz University of Applied Sciences (HSZG). Lab-scale experiments at HZDR and TUD are focused on elucidation of physico-chemical corrosion and precipitation processes as well as resulting clogging effects. Results of generic experiments in a lab-scale corrosion test facility suggest that there is a multi-stage corrosion process. The first stage comprises dissolution of the zinc layer in the coolant forming zinc ions and in turn affecting the coolant chemistry. During the second stage, the base material (steel) corrodes forming insoluble corrosion particles, which can subsequently lead to accelerated clogging of fiber-laden strainers within a few hours. The main influences on corrosion were identified as impact of the coolant jet onto the corroding surface, water chemistry and zinc surface / coolant volume ratio. Furthermore, retrograde solubility of zinc corrosion products in boric acid containing coolants with increasing temperature was observed. Thus, formation and deposition of solid corrosion products cannot be ruled out if zinc containing coolant is heated up during its recirculation into hot downstream components (e.g. hot-spots in core). Corrosion experiments, which included formation of corrosion products at a heated cladding tube, proved that zinc, dissolved in the coolant at low sump temperatures, turns into solid deposits of zinc borates when contacting heated zircaloy surfaces. Due to alternating heating and cooling of the coolant during sump recirculation operation, a cycle of zinc corrosion and zinc borate precipitation may be initiated, which may eventually influence the thermal hydraulics in downstream components during the post-LOCA stage. The results obtained at lab-scale were confirmed by corresponding experiments in semi-technical test facilities of the project partner HSZG. Based on the experimental results, water chemical measures were tested to reduce corrosion and/or zinc borate precipitation effects.

The author recently found that there should exist a “radiation-induced electrolytic (RIE)” mechanism in the reactor water inducing severe interaction between structural materials and their environments in aged LWRs. This mechanism was identified while trying to theoretically reconstruct the potential differences observed in two in-pile test loops; NRI-Rez in Czech Republic and INCA Loop in Sweden. These results are indicating that the in-core potential is approximately 0.1/0.4volt higher, in BWR(NWC)/PWR water chemistry respectively, when compared to the out-core regions. Through modeling studies, it was found that the concentrations of (DH)/(DO) for PWR/BWR(NWC) are higher/lower respectively, in the in-core region compared with the out-of-core region. These solute species in high concentrations should spontaneously decompose at the out-of-core region, enabling control of their water chemistry. This mode of corrosion cell has been dismissed in the nuclear community considering that the transport of ions with flow is insignificant due to high purity of reactor water. Part 1 of this paper focuses on how the RIE phenomena are prompted although the reactor water is kept in high purity. The stable molecular species in the reactor water flow transport the valence electrons. They are released at the cathodic in-core region and are recovered at the anodic out-of-core region. Thus estimated potential differences have been benchmarked with the published in-pile test results for both PWR- and BWR water chemistry environments as explained in Part 2 of this series (1).

The author recently identified that there should exist a “differential radiation cell” mechanism in the reactor water, prompting “radiation-induced electrolytic (RIE)” phenomena. This mechanism was identified while trying to theoretically reconstruct the potential differences observed in two in-pile test loops; NRI-Rez in Czech Republic and INCA Loop in Sweden. Part 2 of this series focuses on the theoretical reconstruction of the observed potential differences. Assuming a state of equilibrium, the author tried to develop a formalism by extending the Nernst equation to reproduce the observed redox potential differences. The radiological potential shift term is separated from the Nernst equation where the latter deals only with stable molecular and ionic species. The radiological effect is described as a perturbation term to the Nernst equation representing a potential shift due to radiation-chemical reactions which should diminish to zero without radiation. The theory generally reproduced the experimental results after fitting the theoretical curve at a single point of the potential for both PWR and BWR-NWC water chemistry environments. This discrepancy is likely due to the “conductive-dielectric property” of the reactor water.

Primary Water Stress Corrosion Cracking (PWSCC) is one of important ageing issues in PWR (Pressurized Water Reactor) primary systems. It has been pointed out that high concentration dissolved hydrogen may lead to occurrence of PWSCC. The authors have proposed to substitute hydrogen by methanol as a fundamental countermeasure of PWSCC. So far corrosion tests of stainless steels and Zircaloy-4 in methanol solutions at 320 °C were conducted under γ-ray irradiation and without irradiation. The test results show that methanol is promising. In the present paper, γ-ray irradiation experiments of methanol solution at 320 °C were done up to 100 kGy. A study on the radiolysis of methanol solution is important from two aspects. One concerns corrosion of structural materials. The radiolysis of methanol may result in formation of harmful compounds to the structural materials, such as carboxylic acids. It is necessary to know the yields of such compounds. The other concern is possible polymerization of methanol and formation of organic polymer deposit on fuel claddings. Large amount of the deposit on fuel claddings should be avoided to keep integrity of fuel claddings. Therefore, it should be clarified whether gaseous species are major products and whether polymerized species of methanol such as ethylene glycol is formed. After the γ-ray irradiation of methanol solution, following species were analyzed: CO 2 and H 2 , methanol, formaldehyde, formate and acetate, and ethylene glycol and glycerin. Without γ-ray irradiation, the major process of the thermal decomposition of methanol at 320 °C is oxidation of methanol by water and generation of one CO 2 molecule and three H 2 molecules. Under γ-ray irradiation, the decomposition of methanol is accelerated; little methanol remains after 10 kGy irradiation. The major product is CO 2 , and polymerization of methanol unlikely occurs. After methanol is completely decomposed, the hydrogen yield still increases. The reducing environment is maintained. Probably, transient organic species play important roles. The addition of low concentration methanol may be sufficient to maintain reducing environment of the PWR primary systems.

This paper updates scientific bases of water chemistry in applying the author’s recent theory, which integrates the elemental radiation- and electro-chemistry reactions in the “Butlar-Volmer equation,” presented in ICONE21-16525. For the past several years the author has been trying to establish that the “long-cell” (a kin to macro-cell) corrosion mechanism is inducing practically all sorts of accelerated corrosion phenomena widely observed in water-cooled reactors, especially in aged plants. The theoretical electrochemical potential differences have been benchmarked with the published in-pile test results for both PWR- and BWR water chemistry environments. However the author’s previous verification efforts were limited to the extent that the curves were fitted with experimental results at a single point. The author re-formulated the basic theory and found that the redox potential difference consists of an electrochemical part (e.g., Nernst equation of dissolved hydrogen or oxygen) and radiation-induced perturbation term, the latter diminishes to zero without radiation. The author continued his studies to clarify whether our current scientific knowledge is sufficient to explain the in-core “chemistry” to reproduce the experimental results without the fitting parameter. Through his study he realized that the basic mechanism of the potential difference is still not sufficiently known. No fitting parameter was used for the PWR water chemistry in the DH region for practical engineering applications, although it is indispensable to confirm the results with an in-pile test loop. In the BWR-NWC the theoretical redox potential out of core was still necessary to be fitted with the experimental results, due to an effect of residual hydrogen peroxide detected by the reference electrode. In addition the calculated potential shift is several times larger than the experimental observation. With the reformulation the scientific validity of the author’s theory is further confirmed. He believes that there is no doubt that the “long-cell” takes place in LWRs, although details are still debatable.

Flow-accelerated corrosion (FAC) is a degradation mechanism that attacks carbon steels under conditions often found in nuclear and fossil power plants. FAC has been responsible for a number of significant accidents in nuclear power plants. The most recent noteworthy accident was at the Mihama Unit 3 (Japan) where the catastrophic failure of a pipe in the condensate system resulted in five fatalities. FAC can affect virtually all of the carbon steel piping and components in the power cycle of nuclear reactors. The presence of wall thinning caused by FAC is determined through the use of non destructive examination (NDE) techniques. For large-bore piping components the most commonly used approach is ultrasonic technique (UT) using an inspection grid applied to the piping components. As FAC is a reasonably well-understood degradation mechanism, a number of computer programs have been developed to help utility engineers determine the inspection locations and managing the data. CHECWORKS™ — the most commonly used computer program for this purpose — is the program discussed in this paper. The use of CHECWORKS™ in utilities’ FAC programs will be described. Particular emphasis will be placed on: inspection planning, handling power uprates and other changes to operating conditions. Inspection planning is one of the most common uses of the CHECWORKS™ software. As there are typically around 5,000 FAC-susceptible components in a reactor system, utility engineers must select the components with the highest FAC-rates for inspection. CHECWORKS™ uses its predictions combined with plant inspection data to provide a best estimate of FAC rates. From these FAC rate predictions and knowledge of the piping schedule and allowable wall thicknesses, inspection locations are determined. Both the water chemistry used and the local operating conditions strongly influence the rate of FAC. CHECWORKS™ is used to study potential changes to water chemistry, system operation or power level to determine the impact of such changes on FAC rates and hence inspection locations. Of particular interest is the use of CHECWORKS™ to determine the impacts of power uprates. Because of the complicated parametric behavior of FAC rates, changing the power level will likely increase the FAC rates in some areas of the plant while other areas will likely see a decrease in FAC rates. This fact requires a pre-uprate analysis to determine how an inspection program will need to be modified. This paper provides a description of how CHECWORKS™ is used in the above applications as well as showing typical examples of its usefulness in these analyses.

When the flaws are detected in Japanese nuclear power components by in-service inspection, structural integrity assessment are performed in the technical judgment on continuous service. If cyclic loading is assumed, fatigue crack growth analysis should be conducted based on the Rules on Fitness-for-Service for Nuclear Power Plants of the Japan Society of Mechanical Engineers Code (JSME FFS Code). However, fatigue crack growth analysis for BWR components consisting of Ni-base alloy is currently impossible, since the reference curve of fatigue crack growth rate for Ni-base alloy in BWR water environment is not yet prescribed in the JSME FFS Code. In this study, fatigue crack growth behavior of Ni-base alloy used for Japanese BWR plants in BWR water environment was investigated. Based on the experimental data, the fatigue crack growth rate curve was evaluated. Four test parameters of material, corrosion potential, stress ratio and load rising time were considered. As a result of fatigue crack growth tests, the effects of all test parameters on the fatigue crack growth behavior were found. A Mean curve of fatigue crack growth rate in Paris law format, which was a function of stress ratio and rising time, was formulated based on crack growth data in normal water chemistry (corrosion potential was over 150 mV SHE ) for weld metal and heat affected zone (HAZ), respectively. A reference curve of fatigue crack growth rate was also formulated by the statistical treatment considering the scatter of crack growth rate. Further, in order to determine the threshold stress intensity factor range Δ K th of reference curve of fatigue crack growth, Δ K decreasing tests were conducted under the test condition of 1 second of rising time. As a result, the threshold value of Δ K was evaluated based on the ASTM E 647, and the Δ K th of the reference curve was conservatively determined considering the margin.

Fibrous debris generated from insulation materials during a Loss of Coolant Accident (LOCA) might be transported to the containment sump strainer. Once the fibrous debris reaches the sump strainer, it might accumulate on the strainer causing loss of Net Positive Suction Head (NPSH), called the upstream effect, or it might penetrate through the strainer becoming a source of clogging of flow channel in the core (in-vessel effect). This is the main idea of the Generic Safety Issue (GSI) 191 initiated by the U.S. Nuclear Regulatory Commission (U.S.NRC). In this study, the authors focused on the quantity of the fibrous debris bypass through a containment sump strainer which eventually might cause the in-vessel effect. In order to produce the fibrous debris, a method suggested by the Nuclear Energy Institute (NEI) was applied to one-side-baked NUKON fiberglass insulation materials. The quantity and characteristics of debris penetrating through the strainer may be affected by the properties of water such as pH and electrical conductivity (EC). Total quantity of the debris bypass for a certain period was measured and the effects of the water chemistry were observed. At higher pH and lower EC more fibrous debris penetrated the strainer.

It has been pointed out that high concentration dissolved hydrogen is one of the important factors of PWSCC (primary water stress corrosion cracking) in the primary systems of pressurized water reactors. Application of a substitution for hydrogen may be a fundamental countermeasure of PWSCC. The authors are developing a new water chemistry technology of a hydrogen alternative to suppress PWSCC. In the present paper, corrosion tests of Zircaloy-4 were performed in deaerated 5 mmol dm −3 methanol solution at 320 °C in the absence and presence of gamma-irradiation. The nominal absorbed dose of the test water was 100 kGy. After the immersion tests, the specimens were analyzed. Weight gain per unit surface area, thickness of oxide film and hydrogen storage were measured. In addition, Raman spectroscopy was carried out, to investigate possible deposition of organic compounds on surface of the specimens. The corrosion behavior of Zircaloy-4 without irradiation agreed with literature data. It was concluded that the presence of methanol did not affect the corrosion behavior of Zircaloy-4. The corrosion behavior of Zircaloy-4 hardly depended on 100 kGy gamma-irradiation. On the Raman spectra of the specimens after the immersion tests, the Raman peaks ascribed to polyethylene or graphite were not found. The deposit of decomposition products of methanol would be negligible if any. It seems that polymerization is not the major process in thermal decomposition and radiolysis of methanol, but methanol decomposes into CO 2 or carboxylic acids.

The interactive effects of prior-deformation and water chemistry on stress corrosion cracking of austenitic alloys in simulated nuclear power plant coolants were quantitatively investigated. Experimental results showed that increasing material yield strength tends to increase stress corrosion cracking growth rates. Increasing electrode potential tends to increase stress corrosion cracking growth rates of austenitic stainless steels. There is a maximum stress corrosion cracking growth rate for Nickel-base alloys and weld metals at electrode potentials near the Ni-NiO equilibrium line. Crack growth rate of prior-deformed austenitic alloys become less dependent on electrode potential than that of their non-deformed counterparts. The modes of prior-deformation and electrode potential affect the stress corrosion cracking path and growth kinetics. The interactive effects between prior deformation and water chemistry on stress corrosion cracking are analyzed.

One key issue during development of Super Critical Water Cooled Reactor (SCWR) is selection of water chemistry operating and control methods for core coolant. Under strong corrosive effect of supercritical water (SCW), the stress corrosion cracking (SCC) sensitivity of materials increases. Most of materials reliably used in PWR or BWR may not be competent for SCWR. Modified austenitic steel and nickel base alloy are hopeful due to low general corrosion rate but failures happen caused by SCC when meet incompatible chemistry environment. The possible options of water chemistry specifications for SCWR should be qualified carefully on SCC effect. This paper introduces research status of SCW chemistry by Nuclear Power Institute of China (NPIC). The SCC behaviors in oxidation and reduction SCW are under study and the testing data will be collected to screen out water chemistry operating strategy for million kilowatts SCWR (CSR1000) under development in China.

In this paper, the author continues his investigation on the scientific basis of water chemistry specifications by applying his recent theory, which integrates the elemental radiation- and electro-chemistry reactions in the “Butlar-Volmer equation.” The B-V equation is well established as the basic material balance equation in corrosion science. The author’s new approach has been compared with the published in-pile test results of the electrochemical potential differences between the in-flux and out-flux regions for both the PWR- and BWR water chemistry environment. Although the theoretical estimation generally reproduced the experimental results, there remains significant deviation from the experimental results at the very low DH region (<10cc-STP/kg-water) in PWRs as well as the low DO region (<10ppb) in BWRs. Although these regions are outside of the water chemistry specifications of general interest, the scientific causes of the deviation must be clarified. In this paper, the author found that the deviations are due to the dominant radiation-chemical reactions involving hydrogen ions and hydrogen peroxide at the lower ends. Although the radiation- and electrochemical reaction was further exploited with respect to the potential differences induced by the hydrogen peroxide, the effects were disappointingly small, when estimated in terms of a mixed potential of the electrode reactions. This leads the author to suspect that hydrogen-ion-radical reactions should be the main causes. Currently further analyses are in progress.

In a series of recently published papers [1,2], the author tried to elucidate on the corrosion environment of LWR cores. Through these studies, the reactor cores are found to be behaving as gigantic cathodes. He established that “long cell action” (a kin to “macro-cell”) corrosion plays a pivotal role in practically all unresolved corrosion issues, irrespective of reactor types and operation [3,4]. The author recently developed a unified theory, which enables the estimating of redox potentials for all types of LWRs (e.g., PWR, BWR, VVER, and RBMK). The author’s previous calculations are confirmed by applying a new set of g -values and reaction kinetics data set [i.e., AECL 153-127160-450-001], developed by Drs. A.J. Elliot and S.M. Bartels in 2009. The results show reasonable agreement with the published in-pile experimental results. However, the author’s calculated DO as well as DH resulted in significant deviations from the input data. The author tried to fortify the verification through another approach in this paper, by incorporating electrolysis reactions (i.e., Faraday’s law) induced by the potential difference. By integrating the radiation-chemical rate equations with the electrolysis reactions, the electric current flowing between the in-core and the out-flux regions through structures was estimated. It was found that this configuration induces surprisingly strong “long cell” current (i.e., cathode current) outside of the regions of DH and DO commonly specified by the water chemistry conditions. This current determines many of the basic corrosion parameters, including ECP, redox potentials, pH, DO and DH, as well as over-potentials, through water chemistry specifications. Although details are still debatable, the author believes that there is no doubt that this corrosion mechanism exists in water-cooled reactors. However it remains debatable due to the lack of basic knowledge pertaining to the chemistry of the LWR cores.

Zirconium alloys are used as structural components in most nuclear reactors. For example Zr-2.5 wt.%Nb alloy is currently used in the fabrication of pressure tubes in CANDU ® (Canada Deuterium Uranium) reactors, and “Zircaloys” are used as fuel cladding in many BWR reactors. In general, zirconium alloys pick up hydrogen (or its isotope deuterium) in reactor primary heat transport system coolant during aqueous corrosion, raising the possibility of hydride formation and delayed hydride cracking at high levels of hydrogen concentration. As a result, the buildup of hydrogen in components due to corrosion could be life limiting. In this paper several mechanisms of in-service hydrogen ingress into zirconium alloys will be discussed to identify the main factors controlling the rate of uptake by the alloy under specific reactor operating conditions. The redistribution of hydrogen due to diffusion will be described along with methods of monitoring and predicting the buildup of hydrogen concentration as a function of operating time. Possible approaches to slow down the rate of ingress will be discussed, which include water chemistry modification and surface conditioning. Innovative non-destructive monitoring techniques developed at “KINECTRICS Inc.” will also be addressed.

Plugging of copper hollow conductors in water cooled generators is an issue occasionally encountered in large nuclear, fossil and hydro powered power plants, causing load limitations or even unplanned shut-down. A response is the proper choice of stator water chemistry and of the purification systems. The common water chemistry regimes are with low or high oxygen content, and neutral or alkaline pH in the stator water. All these regimes have proven to work in the field, but none is without problems. This can be due to various causes. There is one common trait however: problems arise when the system is being exposed for any length of time to a regime it was not designed for. Stainless steel hollow conductors have only simple requirements on water chemistry. Hollow conductor plugging can lead to overheating and possibly to catastrophic failure, so swift action to remedy the situation is recommended. There are several options. Hot reverse flush can remove debris and temporarily open up the plugs, and mechanical cleaning can open totally blocked hollow conductors. There are several methods for chemical cleaning, the most common being different forms of acid cleaning and EDTA cleaning. The former has the disadvantage of dissolving base metal and brazing as well as the oxides, while the latter method manipulates less aggressive substances that only dissolve the copper oxides in the system.

Power plants, including nuclear power plants regularly employ tanks whose contents need to be kept isolated from atmospheric conditions. One way to satisfy these requirements is to provide a liner for the tank which completely fits the interior shape of the tank but floats on top of the tank contents when the tank contains fluid. As the volume in the tank changes, the liner or diaphragm accommodates the changes in volume by sliding along the tank walls. To allow free movement of the diaphragm, management of the gas volume above the fluid and behind the diaphragm is of prime importance. The work described in this paper elaborates on the conditions required to prevent the tank diaphragm from becoming damaged. To develop potential failure modes, the kinematics of the diaphragm and the interaction with the gas volume between the diaphragm and the tank fluid are considered in detail. The developed model is applied to the case of a condensate storage tank at Comanche Peak Nuclear Power Plant (CPNPP). Two physical scale models of the tank were constructed and tested to validate the model and allow the safe operation principles to be quantified for use in the operation of the condensate storage tank at CPNPP. The work allowed CPNPP to design appropriate periodic checks and maintenance activities to ensure the diaphragm will not be damaged due to tank volume changes while still ensuring the required water chemistry criteria for the tank contents can be met.