Keywords

1 Introduction

With the development of nuclear power, more and more spent fuels have been produced from nuclear power plants. China’s policy is to have a closed fuel cycle, where spent fuels are reprocessed for recycling of uranium and plutonium into fresh nuclear fuel and optimizing the management of radioactive waste [1]. In particular, there has been an intensive domestic effort in construction and operation of a commercial reprocessing facility in recent years. Therefore, the safety analysis of reprocessing facility becomes an issue of high priority in the national program to develop spent fuel reprocessing.

Generally, the recognized methods of deterministic analysis are required to be used for safety analysis of reprocessing facilities [2]. In parallel, varying degrees of probabilistic assessments for reprocessing facilities have been carried out in several countries [3–6]. In addition, an Integrated Safety Analysis (ISA) method was developed and applied in the US reprocessing plants (Idaho Chemical Processing Plant and Barnwell plant) [7]. And since 2000, the ISA has been authorized to be an indispensable part of safety analysis for NRC reprocessing facilities [8].

Whatever method of safety analysis is conducted, an identification and systematic analysis of hazard is always the first essential step. Compared with other hazard identification methods such as failure mode and effects analysis (FMEA) and safety checklist (SC), HAZard and OPerability analysis (HAZOP) is suitable to analyze the hazards of complex chemical processes or facilities [5]. It is featured as one of the most suitable methods for performing detailed identification of a wide range of hazards. Recognizing that reprocessing facilities are, to a large extent, chemical processing plants, HAZOP has been logically extended to address radiological and nuclear criticality hazards.

In our research, the HAZOP methodology was introduced to identify and analyze the hazards of a reprocessing facility. Given the limited space available, only part of work on the concentration and denitration process (C/D process) of high-level liquid waste (HLLW) was revealed in this paper, since the C/D process of HLLW has many symbolic features of spent fuel reprocessing, such as high radiation, various chemicals, complex chemical reactions and operation conditions. The purpose of this paper was to test the applicability of HAZOP for a typical process in reprocessing.

Therefore, according to the HAZOP application guide [9], HAZOP analysis was conducted and described in this paper. By a risk matrix, the consequences listed in the HAZOP analysis worksheet were evaluated and categorized into different risk levels.

2 Description of the C/D Process

2.1 Reprocessing and HLLW Management

All over the world, uranium and plutonium in the spent fuel are recovered by a version of the PUREX reprocessing process. The term High Level Liquid Waste (HLLW), generally implies the raffinate (liquid effluent) from the first extraction cycle of reprocessing operations. It contains nitric acid at moderate acidity and greater than 99% of the nonvolatile fission products, almost all minor actinides, together with impurities from cladding materials, corrosion products, several tenths of a percent of originally dissolved plutonium and uranium. Around 5–10 m3 of HLLW is produced per tonne of fuel reprocessed. The HLLW is treated to remove any remaining organic solvents and then concentrated by evaporation to reduce its volume for interim storage in specially designed waste tanks prior to vitrification. Free nitric acid in HLLW is destroyed by reaction with formaldehyde during the concentration process (so-call concentration and denitration process) [10].

2.2 C/D Process

The target concentration after evaporation is corresponding to the equivalent of 110 g/L of fission products oxide in the concentrated HLLW (concentration factor of 6–20 approximately). The final acidity of HLLW is roughly reduced and stabilized to 2–3 mol/L [11].

Generally, the C/D process consists of three major parts as (1) receiving and feeding, (2) concentration and denitration, and (3) off-gas treatment, which are shown in Fig. 1. The first receiving and feeding part is designed to collect HLLW from the first extraction cycle of reprocessing operations, and to transfer HLLW to evaporator. The off-gas treatment part is designed for off-gas decontamination and nitric acid recombination.

In our research, the HAZOP analysis had been done to identify and analyze the hazards of the above three parts. But only the work on the C/D part would be illustrated later.

The major equipment in the C/D part is the evaporator (R-01) which is mainly composed of a kettle type boiler and a bubble-cap-tray decontamination column. The kettle type boiler is heated by superheated water. The stream evaporated from the top of column is condensed through the heat exchanger (E-01). A part of the condensate water is refluxed back to the column to enhance the decontamination performance.

2.2.1 Chemical Reactions

The chemical reactions for denitration between formaldehyde and nitric acid may occur according to three possible reactions as followed [11]:

At high acidity ([HNO3] > 8 mol/L):

$${\text{HNO}}_{3} + {\text{HCHO}} \to 4{\text{NO}}_{2} + {\text{CO}}_{2} + 3{\text{H}}_{2} {\text{O}}$$
(1)

At low acidity (0.5 mol/L < [HNO3] < 8 mol/L):

$${\text{4HNO}}_{3} + 2{\text{HCHO}} \to 2{\text{NO}}_{2} + 2{\text{NO}} + 2{\text{CO}}_{2} + 4{\text{H}}_{2} {\text{O}}$$
(2)
$${\text{4HNO}}_{3} + 3{\text{HCHO}} \to 4{\text{NO}} + 3{\text{CO}}_{2} + 5{\text{H}}_{2} {\text{O}}$$
(3)

Actually, the mechanism of denitration reactions is more complex than the above reactions since the existence of the induction period [12], which refers to the time span needed for nitrous acid to be autocatalytically generated in the mixture to reach a threshold concentration [13]. Because the reactant formaldehyde is added consistently in the induction period without observable reactions with nitric acid, the accumulated formaldehyde may result in uncontrolled runaway reactions later, implying an explosive boiling in the evaporator and accidental release of radioactivity. Research shows that the induction period can be reduced to a few seconds when the concentration of nitrous ions reaching a threshold concentration of about 10–1-10–2 mol/L and by operating at boiling temperature [12].

2.2.2 Three Operation Stages

HLLW is concentrated in the evaporator operated at constant level in a semi-continuous mode which means continuous feeding and batch discharging. The complete sequence of the concentration and denitration procedure is listed below:

  1. a.

    Start-up stage:

    1. 1)

      Feed HLLW into the evaporator (R-01) through L-01 from the receiving tank (V-01).

    2. 2)

      Heat HLLW to the boiling point to start evaporation.

    3. 3)

      Add NaNO2 into the evaporator to reach the concentration of 10–1-10–2 mol/L through L-04.

    4. 4)

      Add formaldehyde into the evaporator through L-05 to start the denitration reaction.

  2. b.

    Normal operation stage:

    1. 5)

      Simultaneously feed HLLW (through L-03) and formaldehyde (through L-05) into the evaporator at a suitable flowrate, keeping the liquid level at constant level. During this stage, the concentration of HLLW increases consistently.

  3. c.

    Shut-down stage:

    1. 6)

      Stop HLLW and formaldehyde feeding when reaching the target concentration of HLLW.

    2. 7)

      Keep heating HLLW with total reflux for a few hours to ensure that formaldehyde has been destroyed.

    3. 8)

      Stop heating HLLW and cool down the evaporator.

    4. 9)

      Transfer the concentrated HLLW to storage tanks.

2.2.3 Potential Difficulties and Monitoring

One major challenge for C/D process is the corrosion risk associated with the acidic solution and high temperature. In addition, the HLLW contains a wide variety of constituents, some of which can promote attack on the stainless steel commonly used for evaporator construction in high acidity. The corrosion risk should be controlled because no direct maintenance operation will ever be possible after the active commissioning.

Reliable monitoring devices are also essentially required for safe operations of the C/D part. The monitor parameters in this part are listed below [11]. The monitoring devices are regarded as a sort of safeguard measurements in hazard analysis.

  1. 1)

    Flowrate of the HCHO and HLLW in L-05 and L-03 respectively.

  2. 2)

    Acidity of the HLLW: acidity of HLLW is measured twice a day by sampling.

  3. 3)

    Liquid level and temperature, pressure inside of the evaporator.

  4. 4)

    Flowrate of the NaNO2.

  5. 5)

    Flowrate of the superheated water and temperature of the superheated water.

Fig. 1.
figure 1

General flow diagram of a HLLW concentration and denitration system [11]

Full size image

3 Methodology and Framework

3.1 HAZOP Methodology

3.1.1 HAZOP Analytical Procedures

HAZOP is a structured and systematic technique for hazard identification. As an inductive tool, HAZOP is often used for identifying a broad range of potential hazards in a system and operability problems likely resulting in nonconforming products.

Generally, the HAZOP analysis process is executed in four phases: (1) definition, (2) preparation, (3) examination, and (4) documentation and follow-up. As the core part of the entire task, the examination phase consists of several major steps as followed.

  1. 1)

    Divide a system into parts, select a part and define design intent.

  2. 2)

    Identify deviation by using guidewords on each element.

  3. 3)

    Identify possible causes and consequences.

  4. 4)

    Identify existing safeguards or protections.

In contrast to HAZOP application guide [9], the step “identify whether a significant problem exists” was excluded from the examination phase. Instead, a risk matrix was established to assess the risks resulted from each consequence/deviation. In addition, some recommendations were proposed for further risk reduction when necessary based on the results of risk assessment.

3.1.2 Elements and Deviations

The selection of elements to be examined is to some extent a subjective decision. For material transferring parts, materials, activities, sources, and destinations can be viewed as elements of the part. For procedural sequence parts, elements may be selected from discrete steps or stages.

The guideword is a specific word or phrase in the HAZOP method used to describe the deviation from design intent. The standard HAZOP guidewords in the process industry include “no”, “more”, “less”, “as well as”, “part of”, “other than” and etc.

Each deviation is then proposed by combining the guideword with the element. Not all combinations will generate credible deviations when all guide word/element combinations are considered. If a credible deviation is identified, it is examined for possible causes and consequences. In our research, causes are only examined in the same part and the consequences can be found in all parts. Then, existing safeguards are also taken into consideration.

3.2 Risk Assessment

In our research, risk is defined as the combination of the likelihood of occurrence of consequence and the severity of that consequence. Risk matrix is a matrix that is used during risk assessment to define the level of risk by considering the category of likelihood against the category of consequence severity. The categories of likelihood of occurrence and severity of consequence were listed in Table 1 and Table 2.

Table 1. Qualitative likelihood classification [14]
Full size table
Table 2. Severity classification [15]
Full size table

A 4 by 5 risk matrix was showed in Fig. 2. There are four different colors – green, yellow, orange, and red – to distinguish the risks according to the likelihood that they will happen and the extent of the damage they would cause. As shown in Table 3, the green zone denotes that the risk is reasonably acceptable, the yellow zone denotes risk that is acceptable with control, and the orange and red zones denote undesirable risk and intolerable risk, respectively [16].

By application of risk matrix, the hazards identified by HAZOP analysis can be further classified. Based on the classified risk, some recommendations were proposed for further risk reduction when necessary.

Fig. 2.
figure 2

Risk matrix [16]

Full size image
Table 3. Risk matrix [16]
Full size table

4 Results and Discussions

Table 4 listed the deviations in the concentration and denitration part. The results of the HAZOP analysis and risk assessment by risk matrix were recorded in the HAZOP analysis worksheet shown in Table 5.

5 Conclusions

The concentration and denitrition process has some symbolic features of spent fuel reprocessing, such as high radiation, various chemicals, complex chemical reactions and operation stages. The HAZOP analysis method was applied in the C/D process. Twenty-two deviations, which were generated by combining elements with guidewords, revealed some potential hazards in the C/D part. After a preliminary risk assessment with the risk matrix, eight of them were recognized as undesirable risks (only accepted when risk reduction is impracticable). The results verified that the HAZOP analysis were suitable for the processes or parts involved with high radiation and complex chemical reactions in reprocessing facilities. More and deeper efforts will be needed in the future to improve the performance of HAZOP for identification and analysis of the hazards in a spent fuel reprocessing facility.

Table 4. Deviations of the concentration and denitration part
Full size table
Table 5. HAZOP analysis worksheet of the C/D part
Full size table