NIH recommends new methodologies for optimizing laboratory hood containment
Editor?s note: The following article is a synopsis of a 520-page study conducted by a project team headed by the author, Dr. Farhad Memarzadeh.
To understand the ways in which complex airflow interactions affect fume hood containment performance in a laboratory, a study was launched by the National Institutes of Health?s Office of Research Services/Division of Engineering Services to analyze more than 250 laboratory configurations.
The study shows how small design and configuration changes affect hood containment performance. For this study, the project team used various configuration parameters, which included: laboratory size, hood position, nominal hood face velocity, supply diffuser type, supply diffuser layout, room ventilation rate, make-up air supply, air temperature, and the presence or absence of a scientist in front of the hood.
The primary fume-hood containment indicators included the following:
1. Sash leakage factor. This factor measures leakage through the sash opening as a fraction of the contamination released inside the hood, which leaks back into the laboratory against the flow. This was found to correlate well with the level of turbulence in the air.
2. Box leakage factor. This factor measures leakage in terms of the escape of contaminants from an imaginary 12-inch box placed just outside the sash opening, which then moves out into the laboratory. This data was found to be a result not only of the turbulence that induces leakage through the sash opening, but also of air currents immediately outside the sash opening.
3. Box/sash leakage proportion. This data measures box performance independently of sash performance. It can also be defined as the proportion of contaminants reaching the box from the sash vs. those emanating from the box into the cleanroom.
Study results
The final report, entitled OMethodology for Optimization of Fume Hood Containment,O contains several recommendations for hood optimization, largely representing new knowledge. It illustrates data by using scatter diagrams of sash leakage factor vs. box/sash leakage proportion and flow diagrams. The data shows the flow from an imaginary particle source, where particles follow the air streamlines and change color according to air speed. After a given time, the particles disappear through the hood, thus preventing the room from filling with particles. The data largely represent new knowledge.
Hood position
Six single-fume hood laboratory configurations were studied, each with the hood placed at the center of the long side wall. The studies were then repeated with the hood located at the corner of the lab against the long wall and at the center of the short wall.
Results indicate that the corner position performs substantially better in terms of both sash and box leakage for a variety of supply air diffusers and laboratory sizes. The only exception is when the hood at the corner is adjacent to the transfer grille. In this case, the jet from the transfer grille passes just above and in front of the sash opening, and then falls down into the working zone of the hood as it mixes with the cool supply air from the diffuser. Because it falls down in front of the open sash, it is detrimental to the box/sash leakage proportion.
Recommendation: Protect the hood by placing it in a corner to avoid jets impinging on the working zone outside the sash opening.
Bulkhead effect
The study also included an examination of how a bulkhead affects airflow. Six single-hood lab configurations, some with a bulkhead and some without, were studied separately. A fume hood cabinet was extended to the ceiling, and then the single-hood configurations were assessed.
Containment was found to vary with lab configuration. When thin, high-velocity jets from a diffuser meet above and in front of the sash opening, the bulkhead forces the jets down directly in front of the open sash, increasing both sash and box leakage. With larger diffusers, the effect is different: only box leakage is affected. In both cases, the resulting jet impacts the air entering the hood, thereby reducing containment.
Recommendation: A bulkhead can reduce leakage when:
1. Using a diffuser layout that gently feeds low velocity air to the hood;
2. Avoiding a diffuser layout that generates high velocity thin jets across the face of the hood from above;
3. Avoiding the use of down-flow diffusers, which causes roll in front of the hood.
Effect of diffuser position
Variation in hood containment performance was tested by moving a 24-inch square diffuser laterally from one side of the hood to the edge of the hood, and then in front of the hood. The overall result in the laboratory was dominated by leakage at the box. The greatest leakage into the bulk of the laboratory occurs when the diffuser is aligned with the center of the hood. Although placing the diffuser in front of the hood reduces sash leakage by approximately 5 percent, the box/sash leakage proportion is increased by about 70 percent, resulting in much larger spread around the lab.
Recommendation: Place a square diffuser on the center line and in front of the hood to minimize sash leakage and minimize exposure to the scientist. However, this does not minimize the leakage into the laboratory.
With the diffuser in front of the hood, leakage was reduced by the presence of a bulkhead. Although moving the diffuser farther away did reduce the sash leakage factor, the improvement was less than 3 percent, a figure constrained by the width of the laboratory. Further, the box/sash leakage proportion increased by almost 12 percent.
Recommendation: Where there is insufficient distance to move the diffuser well away from the hood, position the diffuser in line with the center of the hood close to the bulkhead, so the jet cannot fully develop.
Hood locations
In general, simulations using two hoods demonstrate the extreme difficulty of achieving the same containment levels as with just one hood, when the ventilation rate is dominated by the hood flow rate.
The performance of two hoods on the same wall was simulated using 2-foot to 8-foot separations and a low transfer grille flow rate of 66 cfm. Leakage from the hoods increases when the hoods are close together; however, leakage is different for each hood because of their different positions in the room. When the two hoods are separated by only 2 feet, the highest sash leakage factor and highest box/sash leakage proportion almost doubles as compared to the leakage for a single hood. The highest sash leakage factor is reduced to about 65 percent above that for a single hood when hood separation is increased to 4 or 6 feet. Box/sash leakage proportion for these larger separations is only about 40 percent above that for a single hood. Leakage is further reduced at an 8-foot separation, possibly because the hood is now near a corner.
Containment for hoods on opposite walls is generally better than for hoods on the same wall, except, where two hoods are opposite each other or separated by just 2 feet.
Hoods on perpendicular walls perform best. In addition, they can achieve a box/sash leakage proportion that is lower (by more than half) than that of a single hood, with sash leakage less than 20 percent higher.
Recommendation: Separate hoods by more than four feet. Placing two hoods on perpendicular instead of opposite walls is likely to produce better performance. In turn, either of these configurations can achieve lower leakage than hoods on the same walls. Also, it is best to maximize the distance between the two hoods and the transfer grille.
For further information, write to Dr. Farhad Memarzadeh at National Institutes of Health, Building 15A, Bethesda, MD 20892, or call him at (301) 496-8102, ext. 19.
Dr. Farhad Memarzadeh is a senior mechanical engineer at the National Institutes of Health and the team leader of the fume hood containment study at the NIH?s Office of Research Services/ Division of Engineering Services.
 |
When the hoods are separated by larger distances of 8 feet, the box/sash leakage proportion is only about 40 percent above that for a single hood. Leakage is further reduced at an 8-foot separation, possibly because the hood is now near a corner.
The project teams
The research for this report involved analyzing more than 250 laboratory configurations using Flovent computational fluid dynamics software from Flomerics Ltd., (Surrey, England) and analyzing over 170 million points of data. The research took over 4,800 computer hours (Cray equivalent) and over 4,000 personnel hours.
During the course of the study, several progress meetings were held and the findings discussed. Meetings were attended by some 50 key personnel, covering the full spectrum of laboratory design and operation, including architects, engineers, industrial hygienists, scientists, and experimentalists from a multitude of organizations as well as from the private sector. – FM