Vapor fences

Recent work done on physical vapor barriers was performed by the Industrial Cooperative HF Mitigation/Assessment Program (ICHMAP) (Petersen and Radcliff, 1989). This program studied the effects of vapor fences and vapor boxes. The primary objective of the study was to determine the effectiveness of these devices to retard the transport and to dilute heavier-than-air (HTA) releases of a toxic material like hydrofluoric acid (HF). Because vapor barriers could also see releases of flammable materials, an effort was made to determine their impact on the consequences of a vapor cloud explosion. 

Vapor fences are solid walls located downstream from an expected release point of a flammable or toxic gas. These fences are most effective in achieving an initial dilution when they are located as close to the source as possible but downwind of it, or on the side of a facility where a sensitive population is located. Heights of vapor fences can vary from 3 to 12 m. At times they are added to the tops of dikes to provide a dilution effect (see Section 5.2.1). 

A vapor fence works by forcing toxic or flammable vapor to pile up behind it, then reach the fence height, and finally flow over the top of the fence. When the vapor begins to flow over the fence, it becomes entrained in the wind and dilution occurs. Until flow over the top occurs, the fence acts as a barrier with only small amounts of material being carried downwind. 

Fences can assume various configurations. The simplest is a straight barrier. A more sophisticated straight vapor fence consists of several components of various lengths placed one behind the other with gaps between them. Another type of vapor fence forms a semicircle of a constant radius from the expected release point. Other configurations are possible and should be selected based on modeling studies (Meroney, 1991). 

The interaction of dispersing clouds with vapor fences is a complex physical process. When a flow meets an obstruction, turbulence levels are increased downstream because of vorticities introduced into the flow field, and increased velocity gradients are induced by flow momentum losses. Detailed modeling of such a process is very difficult and requires a combination of small-scale experiments and computational fluid dynamics. 

---

Vapor barriers or vapor fences to contain dense, evaporated vapors and promote dispersion and... 

A simple concept can be used to illustrate important features of vapor barriers that alter cloud dispersion behavior in the near field (Meroney, 1991 Meroney and Neff, 1985). The concept is based on adding an entrainment velocity ( e) contribution, which is attributed to the vapor fence. 

Meroney (1991) and Meroney and Neff (1985) proposed a simple equation (5.1) to predict the increase in air entrainment by a vapor cloud caused by the presence of a vapor fence ... 

Figure 5.5. Vapor fence effects on downwind concentration profile (Meroney, 1991).

Assessing the effects of enclosures or vapor boxes is similar to assessing the effects of vapor fences. Consider the case of a typical dike sized to hold 110 percent of the liquid spilled. For small releases, the dike walls would act as containment or storage for the vapor generated shortly after release occurred. The vapor that would be generated by evaporation or boiling from the dike floor would displace the air in the dike for a given period of time and then overflow the dike walls. The vapor holdup duration is easily estimated ... 

Figure 5.7. A leaky vapor fence (Arthur D. Little, Inc., 1974).

Van Zele and Diener (1990) state that vapor fences can reduce the near-field concentrations of hazardous materials, but that concentrations increase with increasing downwind distance. Wind tunnel data show that vapor fences will reduce near-field concentrations by factors of 2 to 9, but the concentrations downwind at approximately 1000 m from the release point will eventually equal concentrations when no fence is in place. For a heavier-than-air cloud, there is no appreciable delay in the cloud s arrival time. 

Description of Experiments The Falcon Series of liquefied natural gas (LNG) spill experiments was performed by LLNL during the summer of 1987. These were the first tests performed at DOE s permanent Spill Test Facility. A series of five spills was performed on water within a vapor barrier structure as part of a joint government/industry study. These experiments were performed to evaluate the effectiveness of vapor fences as a mitigation technique for accidental releases of LNG. They also provided a database for the validation of wind tunnel and computer modeling simulations of vapor fence effects on LNG dispersion. Spills were made onto a water pond equipped with a circulation system to maximize evaporation to make the source evaporation rate as nearly equal to the spill rate as possible. 

---