A. SANAD and Y. ALOUFI, Saudi Aramco, Dhahran, Saudi Arabia; and P. DHOTE and A. TAWATE, Siemens Energy, Abu Dhabi, United Arab Emirates
In the context of industrial risk management, the height of the flare stack and the reliability of pilot control systems are crucial factors in the design and operation of industrial facilities. When identified risks associated with toxic and flammable dispersion are deemed unmanageable, the proactive consideration of engineering solutions at the early stages of design becomes imperative to mitigate these risks and enhance safety measures.
Elevating flare stack height represents a significant engineering intervention aimed at minimizing the potential impact of toxic and flammable clouds reaching ground level. By increasing the stack’s height, the dispersion and dilution of pollutants into the atmosphere are optimized, thereby reducing the likelihood of toxic and flammable clouds coming into contact with the surrounding environment. This approach not only aligns with regulatory requirements for minimizing ground-level impacts but also significantly contributes to overall environmental and public health protection.
Moreover, the reliability and integrity of pilot control systems play a paramount role in preventing unintended flameout events in flare stacks. Enhancing the reliability of these systems through advanced engineering and technology serves to significantly minimize the risk of flameout incidents leading to the release of unburned pollutants. This proactive measure not only safeguards against potential environmental and safety hazards, but also contributes to the overall operational efficiency and risk management of industrial processes.
By integrating these engineering interventions at the early stages of design, industries can effectively manage and mitigate the potential consequences of toxic and flammable release scenarios. This proactive approach aligns with the central theme of this article’s emphasis on analyzing dispersion and selecting optimal scenarios for analysis, emphasizing the critical role of engineering solutions in addressing risks associated with toxic and flammable dispersion, and enhancing overall safety and environmental protection.
Dispersion analysis. Dispersion is the process by which a substance, such as a toxic chemical or pollutant, is spread and diluted in the surrounding environment. In the context of industrial processes, dispersion analysis is crucial for understanding how these substances disperse into the atmosphere and potentially impact surrounding areas.
One key aspect of dispersion analysis is the consideration of flare stacks. Flare stacks are commonly used in industrial facilities to burn off waste gases and prevent their release into the atmosphere. However, in the event of a failure or malfunction, these flare stacks can release pollutants and toxins into the air, posing significant environmental and health risks. Dispersion analysis helps to predict how these pollutants will spread and disperse, aiding in emergency response planning and risk assessment.
Flammable dispersion, particularly flameout dispersion, refers to the dispersion of pollutants resulting from the extinguishing of a flare stack flame. When a flare stack flame goes out unexpectedly, there is a risk of unburned hydrocarbons and other pollutants being released into the atmosphere. Understanding the dispersion of these pollutants is crucial for evaluating the potential impact and implementing effective mitigation measures.
The first step in dispersion analysis is selecting the appropriate toxic and flammable scenarios from different emergency scenarios or from routine or non-routine flaring. The import factors in selecting the appropriate scenarios are listed here:
Risk assessment: Different emergency scenarios, routine flaring and non-routine flaring events can lead to the release of various types and quantities of toxic and flammable substances. By selecting and analyzing these scenarios, it becomes possible to assess the potential risks associated with each type of release. This information is essential for developing effective emergency response plans and for implementing appropriate risk mitigation measures.
Environmental impact: Understanding the dispersion patterns and potential scenarios for toxic and flammable releases from different emergency situations and flaring events is essential for evaluating their environmental impact. By selecting the most relevant and realistic scenarios, it becomes possible to predict the potential environmental consequences of toxic and flammable dispersion, aiding in the development of strategies to minimize environmental damage.
Health and safety considerations: Different toxic and flammable release scenarios can have varying impacts on human health and safety. Selecting the appropriate scenarios allows for the assessment of potential exposure risks to workers, nearby communities and the public. This information is crucial for implementing measures to protect human health and safety in the event of a toxic and flammable release.
Regulatory compliance: Many industries are subject to regulatory requirements regarding the management of toxic and flammable substances and the prevention of their release into the environment. By selecting and analyzing the most relevant toxic and flammable scenarios, companies can ensure compliance with regulations and standards governing toxic and flammable dispersion and emergency response.
In summary, the importance of selecting the appropriate toxic and flammable scenarios from different emergency scenarios or from routine and non-routine flaring lies in the ability to assess risks, evaluate environmental impacts, protect human health and safety, and ensure regulatory compliance. This selection process is essential for developing effective strategies to manage and mitigate the dispersion of toxic and flammable substances.
In the context of managing the risks associated with toxic and flammable dispersion, the consideration of stack height and the reliability of pilot control systems plays a crucial role in the design and operation of industrial facilities. If the identified risks from toxic and flammable dispersion are deemed unmanageable, proactive engineering solutions can be considered at the early stages of design to mitigate these risks and enhance safety measures.
Increasing the stack height is a significant engineering intervention aimed at minimizing the potential impact of toxic and flammable clouds reaching ground level. By elevating the stack height, the dispersion and dilution of pollutants into the atmosphere are enhanced, reducing the likelihood of toxic and flammable clouds coming into contact with the surrounding environment. This approach not only aligns with regulatory requirements for minimizing ground-level impacts but also contributes to the overall environmental and public health protection.
In a subsequent case study, an assessment was conducted on an existing site flare stack to analyze flameout scenarios for both toxic and flammable substances at varying flowrates. The primary objective was to determine whether the dispersion plume was reaching ground level or if it was effectively being dispersed and diluted into the atmosphere. By simulating different flowrates and evaluating the stack height in relation to the reliability of pilot control systems, the study aimed to identify potential risks and assess the effectiveness of existing safety measures. The findings from this case study provided valuable insights into the behavior of toxic and flammable dispersion and helped determine the need for proactive engineering solutions to mitigate risks and enhance safety measures at the industrial facility. This detailed analysis contributed to a better understanding of the impact of stack height and pilot control systems on the dispersion of hazardous substances, ultimately leading to informed decision-making that improved environmental and public health protection.
CASE STUDY: CONTROLLING SCENARIOS FOR DISPERSION OF GAS PROCESSING FACILITIES
Flare and relief system design, including radiation analysis, is typically dictated by the maximum instantaneous relief load. However, in dispersion analysis for flameout scenarios, the design rate in the dispersion analysis may not be same as the flare sizing case. Dispersion studies are sensitive to high toxicity and flammability mixtures of certain components in the flared gas. To accurately account for system dynamics, this article will discuss ways to identify and capture all the critical process releases that may impact dispersion studies for flare design. Additionally, a dedicated case study will be demonstrated in the article.
The instantaneous flow is defined as the maximum sum of the loads resulting from any single contingency (including the global scenarios) considered for emergency relief and/or emergency shutdown requiring depressurization, and the plant(s) or unit(s) feed gas flow associated with that relief or shutdown situation. Such maximum load is used to size and determine the size and configuration of the flare tip relief headers, radiation analysis and others. Although a maximum instantaneous scenario could be the controlling scenario for dispersion, a conceptual mistake is common when lacking more critical controlling scenarios for dispersion analysis.
FIG. 1 shows a typical gas sweetening facility, where feed gas is routed to the slug catchers for physical separation. Therefore, the separated gas is routed further to feed gas separators to avoid solid carryover to the system. The feed gas is then sent to the absorption-regeneration process. The acid gas will be observed by the amine system to enable proper sweetening for the downstream facilities. The rich amine is regenerated through the stripper column where acid gas is stripped, and lean amine is recycled back to the process for a new cycle of absorption through the circulation pumps.
In such processes, the maximum instantaneous load will be due to the blocked outlet of the amine contactor or emergency shutdown trigger due to the outage of the circulation pump (amine) to avoid sending sour gas to the downstream facilities. The resulted event will lead to huge flaring that could equal the train’s process gas capacity (e.g., 450 MMft3d). However, the compositions of the relief scenarios are usually light components such as methane and ethane. Hence, such components will result in a light cloud that would have good dispersion. The other large load that would be a sizing scenario for the acid flare is gas blow-by from the amine contactor to the amine flash drum. However, this scenario would result in a rich hydrocarbon.
Therefore, a critical scenario in this case could be the blocked discharge on the acid gas that is sent to the sulfur recovery unit (SRU). The resulted relief would be a low volume, but the concentration of the acid would be high. In fact, such scenarios are more likely to result in a flameout due to the existence of carbon dioxide (CO2), which drops the heating value of the stream. Hence, in such a process, assist gas is usually provided to maintain the proper British thermal unit (Btu) value. The dispersion of such a scenario could end with a critical situation where the acid gas would be heavier than the air and could drop to personnel/ignition source levels. Therefore, the level should include the ground level and any other levels (e.g., nearby platforms). Likewise, the flaring of liquefied sweet gas in some facilities located downstream is another critical case. In such a process, some of the flaring events could end with a two-phase flow with much heavier compositions.
A case study was conducted at an existing facility to conduct dispersion modeling using various sensitivity cases to assess the conditions under which a toxic cloud might reach the ground. From the numerous cases analyzed, two specific scenarios were selected for in-depth investigation:
Gas blow-by on an amine flash drum
A blocked outlet on the acid gas discharge to the SRU.
For these selected scenarios, the flare load was systematically varied to account for different weather conditions and Pasquill stability classes. The input parameters for the dispersion study, as well as the basis and conclusions derived from the study, are outlined below.
The selected flare system was an acid flare. The original analysis indicated that the toxic dispersion cloud was not touching the ground. Therefore, the following approach was used for the case study.
Scenario selection. The case study’s scenario selection included:
Listing all the emergency, routine and non-routine scenarios relieving to the flare header
Checking the off-spec, startup and shutdown philosophies and other modes of operations
Identifying the relief compositions for each flaring scenario
Dividing the scenarios between flammable and toxic release (a few scenarios might fall into both categories)
Screening high flowrates and high molecular weight streams with flammable components
Selecting the cases where hydrogen sulfide (H2S) or other toxic components were present
Selecting the cases where composition results were in the low Btu values (< 500 Btu/ft3)
Screening the selected scenarios further with high toxic component mass fraction, high molecular weight stream and high flowrates
Selecting scenarios for flammable and toxic dispersion analysis should account for varying flowrates to encompass any reductions in flow over the duration of the scenario.
Setting up the dispersion model. Input data for the dispersion model is detailed in TABLE 1. TABLES 2–4 detail the scenarios and relief loads considered for the flameout. Note: The screening of all individual scenarios carried out in the case study was compared against global scenarios to ensure that the selected case was conservative.
Methodology to estimate the flare tip exit velocity. The following steps were taken to estimate the flare tip exit velocity:
Step 1: Using the flare modeling applicationb, a model was developed for each of the flow cases, namely 100%, 75%, 50%, 25% and 1%.
Step 2: The flare tip inlet conditions obtained from the respective relief flow case file were used as input in proprietary softwarea to estimate the flare tip exit velocity and exit temperature.
Step 3: Using the proprietary softwarea, the relief fluid was defined with an appropriate property package to estimate the thermodynamic properties of the mixture.
Step 4: In the softwarea, the flare tip was defined with a vendor-provided capacity velocity pressure drop curve.
Step 5: The flare tip exit velocity and exit temperature obtained from the softwarea were used in PHAST dispersion modeling.
TABLES 5 and 6 detail estimated exit velocities at the flare tip for Cases 1 and 2.
Case 1: Results of dispersion modeling. The following are the flammability and H2S toxicity results of the dispersion modeling for Case 1.
Flammability. The flammability case was considered for 0.5 LFL, indicating that the dispersion analysis accounted for the potential spread and impact of flammable substances at concentrations near the lower flammability limit. FIGS. 2–6 represent the dispersion plume for different flowrates and weather conditions.
H2S toxicity. FIGS. 7–16 represent the dispersion plume of H2S toxicity for different flowrates and weather conditions.
Case 1: Summary of results. The results for Case 1 are summarized in TABLE 7.
Case 2: Results of dispersion modeling. The following are the flammability and H2S toxicity results of the dispersion modeling for Case 2.
Flammability. The flammability case was considered for 0.5 LFL, indicating that the dispersion analysis accounted for the potential spread and impact of flammable substances at concentrations near the lower flammability limit. FIGS. 17–21 represent the dispersion plume for different flowrates and weather conditions.
H2S toxicity. FIGS. 22–31 represent the dispersion plume of H2S toxicity for different flowrates and weather conditions.
Case 2: Summary of results. The results for Case 2 are summarized in TABLE 8.
Sensitivity analysis, considering the impact of assist gas. Cases where the plume was touching the ground were re-evaluated, considering the flow of assist gas. Assist gas ensures proper combustion and mitigates the likelihood of flameout scenarios.
A flow of 30-MMft3d assist gas (fuel gas) was combined with the relief stream to obtain the resulting composition and flowrate. The exit velocity at this resulting flowrate was then determined using the methodology described earlier. It was observed that due to the diluted composition, the plume was no longer touching the ground. The results of this observation are shown in FIGS. 32–34.
Based on FIGS. 32–34, it is evident that the assist gas can effectively dilute the composition and prevent the plume from touching the ground. However, the effectiveness of the assist gas depends on the ratio of the flare gas flowrate to the assist gas flowrate. In scenarios where the flare gas flowrate is significantly higher, the impact of the assist gas flowrate may be limited.
Takeaways and recommendations. The results of the dispersion analyses for both flammable and toxic fluids highlight the critical importance of understanding and evaluating the potential impacts of release scenarios in industrial settings. The dispersion modeling has provided valuable insights into the behavior of flammable and toxic clouds, indicating their maximum downwind distances and potential contact with the ground under varying relief flowrates and concentrations.
Flammable fluid dispersion. The dispersion analysis for the flammable cloud demonstrates that, in both cases, the cloud does not touch the ground; however, it disperses to significant downwind distances. In Case 1, it reached a maximum distance of 360 ft at 100% relief flow, with Case 2 showing a maximum disbursement distance of 37 ft at 25% relief flow. These findings emphasize the importance of understanding the potential reach of flammable clouds, and the need for effective risk management strategies to prevent ignition and mitigate potential hazards.
Toxic fluid dispersion. The dispersion analysis for the toxic cloud revealed varying outcomes for different scenarios. In Case 1, the toxic cloud does not touch the ground, dispersing to maximum distances of 545 ft for a 30-ppm H2S concentration at 100% relief flow and 104 ft for a 100-ppm H2S concentration at 25% relief flow. However, in Case 2, the 30-ppm toxic cloud touches the ground at 100% and 75% relief flow, while the 100-ppm toxic cloud touches the ground at 100% relief flow. Additionally, the toxic cloud disperses to extensive downwind distances, reaching a maximum of 9,200 ft for a 30-ppm H2S concentration at 100% relief flow and 3,320 ft for a 100-ppm H2S concentration at 25% relief flow. Note: Some fluid incorporates with high pressures, two-phase flow or chemicals that should require a further in-depth analysis of hydrate formation, freezing and others that could be discussed in a future different case study.
Wind direction. Prevailing wind is an important basis for the flare location selection due to the dispersion analysis. Typically, the flare should be located crosswind to avoid flammable or toxic gases blowing directly towards personnel or ignition sources. Nevertheless, the wind direction in the modeling should consider the worst-case scenarios toward the reference points.
One important aspect is to evaluate the dispersion from and to the flare tips, including ground and horizontal flares. This should also include nearby flare tips, evaluating the potential of explosions of the disposed cloud.
Determine the dispersion’s worst-case scenarios. To determine worst-case scenarios, consider the most critical flaring scenarios in terms of compositions, pressure, temperature and load (FIG. 35). The compositions selected should consider any combination that could lead to toxic, flammable or burning rain. As described earlier, in several scenarios, a local scenario with a low volume could be more critical than high-volume scenarios. That is due to the proportional impact between the molecular weight and dispersion analysis. Therefore, flowrates are not as critical as the compositions set. Moreover, in some scenarios, the compositions could end with a two-phase flow, which may end with burning rain. Such scenarios should be evaluated to avoid the impact on burning rain in the area.
Source level for the dispersion analysis. The concept of the source level in dispersion analysis is missed in several studies. In many dispersion analyses, the affected areas considered are only ground levels. However, the dispersion should also consider nearby piping, platforms, equipment, adjacent flares and other accessible areas. In several facilities, a partial shutdown could take place. In such scenarios, even a partial shutdown of flare systems is expected.
Furthermore, the impact of dispersion analysis includes all the accessible platforms that may not necessarily be on ground level. Site verification of any dispersion analysis is a key activity to ensure adequacy and safety.
Recommendations. Based on the dispersion analysis results, the following recommendations were proposed:
Enhanced emergency response planning: Given the potential for toxic clouds to reach significant downwind distances and touch the ground in certain scenarios, it is crucial to enhance emergency response planning to effectively manage and mitigate the impact of toxic releases. This may include the development of specific response protocols, and the implementation of proactive measures to minimize human exposure and environmental damage.
Flare system optimization: Considering the varying dispersion outcomes for different relief flowrates, there is a need to optimize flare system operations to ensure the effective control and management of flammable and toxic releases. This may involve the implementation of advanced control systems and technologies to mitigate potential dispersion risks and improve overall system reliability.
Regulatory compliance and reporting: The dispersion analysis results highlight the importance of aligning with regulatory requirements related to the management of flammable and toxic substances. It is recommended to ensure compliance with relevant regulations and standards governing dispersion management, as well as to establish robust reporting mechanisms for monitoring and addressing potential dispersion events.
The dispersion analysis should determine the impact on adjacent flares, accessible platforms and the source of ignitions, among others.
Conduct a site verification to assess the impact of the dispersion analysis in all cloud plume directions and levels. A 3D model would be sufficient for grassroots facilities.
Consider providing assist gas in an automated mechanism with an advanced control arrangement to reduce the potential of a flameout scenario and the consequences of any flameout scenarios, as illustrated in FIGS. 33 and 34.
In summary, the dispersion analysis findings underscore the significance of proactive risk management, and the implementation of effective strategies to address the potential impacts of both flammable and toxic dispersion in industrial facilities. These recommendations aim to enhance safety, environmental protection and regulatory compliance in the management of dispersion-related risks.
Disclaimer. The information contained in this article represents the current view of the authors at the time of publication. Process safety management is complex, and this document cannot embody all possible scenarios or solutions related to compliance. This document contains examples for illustration and is for informational purposes only. Saudi Aramco and Siemens Energy make no warranties, express or implied, in this article. Furthermore, any adverse impact resulting from the implementation of the solutions provided in this document may not be attributed to the authors or their respective companies. HP
NOTE
SLB’s Flaresim
Aspen Technology’s Flarenet
LITERATURE CITED
National Institute for Occupational Safety and Health (NIOSH), “Hydrogen sulfide,” U.S. Center for Disease Control and Prevention, online: https://www.cdc.gov/niosh/idlh/7783064.html
Abdullmajeed I. Al Sanad is a Process Engineer at Saudi Aramco specializing in the flare and relief systems field. Al Sanad is works as Project Engineer at North & South Gas Increment Development. He has 9 yr of experience that includes company operations, project support and technology development. Al Sanad has also led major efforts in Saudi Aramco flaring minimization programs. He holds more than ten patents related to emissions monitoring, flare gas recovery systems, flare integrity and ignition systems and earned a BS degree in chemical engineering from King Fahd University of Petroleum and Minerals in Saudi Arabia.
Yousef D. Aloufi is a Process Engineer at Saudi Aramco, working as a subject matter expert in flare and relief systems. He has led key initiatives to reduce the environmental impact of industrial processes. His contribution in developing innovative flare monitoring systems and minimization plans has resulted in substantial reductions in both emissions and costs. Aloufi holds five patents related to emissions monitoring in flare systems. Additionally, he has played a critical role in designing and optimizing flare and relief systems for major gas, oil and refining projects. Aloufi earned a BS degree in chemical engineering from the University of New Brunswick, and an MS degree in chemical engineering and an MBA from Cornell University.
Praveen Dhote is a Process Safety Team Leader at Siemens Energy. Dhote earned a BTech degree in chemical engineering. He has 23 yr of experience specializing in process safety management. He graduated from Laxminarayan Innovation Technological (LIT) University in Nagpur, India. He has extensive experience in pressure relief and flare system design and revalidation. Dhote is a certified functional safety engineer, chairs HAZOP and SIL workshops, and is a certified Hydrogen Safety Engineer from Ulster University, Ireland.
Avinash Tawate is a Project Technical Lead at Siemens Energy. He has 22 yr of experience in process engineering and process safety management. He has experience in PDP, FEED and EPC-phase projects for the refining, petrochemicals, specialty chemicals, and oil and gas industries. Tawate is a certified Value (Engineer) Methodology Associate, certified by SAVE International.