Grease Extraction - Myth and Reality •
Grease Defined by Compound Elements
•
EPA Test Method 5
•
Weighted Filter Test Methods
•
Grease Extraction Efficiency Test Protocol (VDI Standards)
•
Cyclonic Grease Extraction
•
Ultra Violet Light Technology
Cooking in commercial kitchens can produce large amounts of heat, water vapor, smoke, and perhaps most significantly grease. The amount of grease produced by cooking is a function of many variables including: the type of appliance used for cooking, the temperature that food is being cooked at, and the type of food product being cooked. The purpose of a mechanical grease filter is to twofold: first to provide fire protection by preventing flames from entering the exhaust hood and ductwork, and secondly to provide a means of removing large grease particles from the exhaust stream. The more grease that can be extracted the longer the exhaust duct and fan stay clean, resulting in better fire safety. From a practical standpoint, grease filters should be easily cleanable and non-cloggable. If the filter becomes clogged in use, the pressure drop across the filter will increase and the exhaust airflow will be lower than designed. In terms of grease removal efficiency, many kitchen hood manufacturers would claim that the efficiency of their grease filters is 90 % or higher. They would reference to UL 1046, ULC-S64993 or Navy NBSIR 74-505 standards to support the performance measurements. Unfortunately none of these standards represent filter performance under cooking conditions. Grease emissions from cooking processes consist of both particulate and vapor. Mechanical grease extractors can be effective only at capturing particulate and are not able to extract vapor. That is why, theoretically, the maximum grease extractor efficiency can not exceed the mass fraction of particulates in the total cooking emissions. This maximum efficiency depends on cooking operations and ranges from 2 to 70% for electric ovens and gas broilers, respectively, based on the percentage of particulate emissions from the cooking process. Return to the top of the page
Grease Extraction - Myth and Reality Grease Defined by Compound Elements Click here to return to the Grease Extraction Page
According to the University of Minnesota (Gerstler, et. al.) ,grease is comprised of a variety of compounds including solid and/or liquid grease particles, grease and water vapors, and a variety of non-condensable gases including nitrogen oxides, carbon dioxide, and carbon monoxide. The composition of grease becomes more complex to quantify as grease vapors may cool down in the exhaust stream and condense into grease particles. In addition to these compounds, hydrocarbons can also be generated during the cooking process and are defined by several different terminologies including VOC (volatile organic compounds), SVOC (semivolatile organic compounds), ROC (reactive organic compounds) and many other categories. The amount of grease emitted during the cooking process is dependent on several variables; let us look at these in more detail.
Return to the top of the page
Grease Extraction - Myth and Reality EPA Test Method 5 Click here to return to the Grease Extraction Page
Historically, the most commonly used method to quantify the mass of grease that a grease extractor removed was to use the U.S. Environmental Protection Agency’s (EPA) Method 5 test. Several derivatives of this method have been implemented by testing agencies around the country, the most recognized of which is the method used by the South Coast Air Quality Management District (SCAQMD) in Los Angeles. These EPA Method 5 tests measures the particulate and vapor mass that bypasses the filter while the SCAQMD protocol adds the measurement of volatile organic compounds (VOC) to the basic procedure. The basic test setup is shown below in Figure 22.
In application the probe is places inside the exhaust duct and the temperature of the probe is allowed to stabilize. The cooking process is then started and the grease particles and vapor are entrained into the sampling train. Inside the filter holder is a 0.3 micron glass fiber filter that removes the particulate matter from the entrained air, or alternatively a particle size impactor may be used at this location to further separate the grease particles into discrete size ranges. Next, the entrained exhaust stream enters a series of impingers, placed in an ice bath to cool them, and condensable vapors are collected at this location. Finally, volatile organic compounds may be analyzed using a hydrocarbon analyzer on the remaining entrained exhaust air. The results of this test method are generated by performing a mass analysis on one more glass fiber filters and a set of liquid impingers. Prior to testing, the filter(s) are desiccated till dry and pre-weighed and then desiccated and post-weighed after testing is completed. This difference in mass is the weight of particulate matter for the particle size range collected that was produced by the cooking process and is normalized to the amount of food that was cooked during the sampling period. A cascade impactor is commonly used to collect samples when multiple particle size cuts are desired. One drawback with the cascade impactor is that the filter for each substrate needs to be weighed before and after the analysis which means there are more chances for error with the measurements. Depending on the number of stages in the cascade impactor, there might only be two or three stages that collect particles in the size range of 2 to 10 microns which is the range that differentiates filter performance among competing products. The impingers are cleaned prior to testing and rinsed with Acetone after testing is completed and are placed in pre-weighed dishes to evaporate. After the evaporation is completed, the amount of mass gain in the dishes is equal to the mass of condensable vapors emitted by the cooking process. Once again, this is typically normalized to the amount of food cooked during the sampling period. Another problem with the EPA method 5 tests is that it was designed for sampling in industrial and power plant stacks, not to measure the comparatively small amounts of grease produced by cooking processes. Secondly, it can take several days after testing to obtain the results due to the need to dry the filters and wait for the impinger liquid to evaporate. The added time also increases the costs of performing the EPA Method 5 test.
When using EPA method 5 in combination with cooking real food products, the repeatability of the grease emissions is a concern. When cooking burgers on a broiler, there can be a variation of 20% in the grease emissions due to differences in the meat mixture and amount of fat in the individual hamburger patties; even though the batch of burgers is mixed to provide a certain fat content overall.
Return to the top of the page
Grease Extraction - Myth and Reality Weighted Filter Test Methods Click here to return to the Grease Extraction Page
Click here to download the Filter Comparison test Charts in Excel format 19kb
A study on filter efficiency was conducted by Halton on a broad variety of filters (Livchak, et. al., 2003). Using the principles of VDI standard 2052, filters from different manufacturers were selected to be tested for extraction efficiency. Halton’s KSA filter, a generic baffle filter, and a top-rated baffle were tested (Figure 5). The measurements were conducted at VTT (Technical Research Center of Finland, an independent state-owned research organization). Efficiency measurements were made by using DOS (di-ethyl-hexyl-sebacate) particles generated with a Spinning Top Aerosol Generator in a particle size range of 4 – 10 μm and with a pneumatic nebulizer in the particle size range of 1 – 5 μm. Particle concentrations from the upstream and downstream of the grease filter were measured with an aerodynamic particle sizer (APS). The filters were placed in a test stand at an angle of 45° and test particles were introduced into a mixing chamber. Filtered supply air was directed into the mixing chamber and then to the test stand. The operation of the spinning top aerosol generator was controlled during the measurement so that particles up to 10 μm were generated.
Measurement Results Measurement results were made at approximately 200 cfm/ft of filter width (see Figure 6). By comparing the performance at the same airflows, this reflects the actual filter application in the field.
Halton’s KSA Filter
Generic Baffle Filter
Top-Rated Baffle Filter (All filters are 19.5” (495 mm) long
Figure 6. Filter Efficiency as a Function of Particle Size for Three Filters. As one can see from Figure 6, none of the mechanical filters are effective in capturing particles less than 2.5 μm in size. Filtration efficiency varies significantly with the filter design and manufacturer. Halton’s KSA filter traps at least twice the amount of grease compared to the top-rated baffle filter in the particle range from 2.5 to 10 μm at the exhaust airflow of 200 cfm/ft. It is also interesting to point out that filtration efficiency for all filters increases with the airflow as a result of higher pressure drop across the filter. Applying the Filter Efficiency to Cooking Processes By knowing the efficiency of filters as a function of particle size and particulate distribution for different cooking operations, it becomes possible to calculate filtration efficiency for different cooking operations. The method consists of three stages: 1. Determine emissions from a cooking process (vapor and particulate) including particle distribution for particulate emissions. 2. Determine the fractional filter efficiency as a function of particle size and range of airflows through the filter. 3. Calculate the filter efficiency for the cooking process by overlaying the filter efficiency curve with the particle distribution curve for the cooking process at the design (for this cooking operation) airflow through the filter.
The benefit of this method is that stage 1 and 2 are conducted independently and there is no need to perform filter efficiency test for every cooking process (as would be the case with EPA Method 5) as long as there is adequate information on the emissions from the cooking process (vapor, particulate and particle distribution) and the filter efficiency as function of a particle size. Calculation results of the filters efficiency as a function of some typical cooking process are presented in Figure 7 and Figure 8. Two filter efficiencies are presented: total, as ratio of grease captured to total cooking emissions and particulate efficiency as ratio of grease captured to particulate emissions. Even for the best grease extractors, the efficiency does not 90% due to the fact that mechanical filters cannot trap vapor effluents.
Cooking Operation
G as B ro iler H am b urg er
G as F ry er
G as G rid d le
0%
20% 40% 60% 80% Total Grease Removal Efficiency (%) Top-R ated B af f le
G eneric B af f le
100%
H alton K S A
Figure 7. Total Grease Removal Efficiency for Three Filters With Different Cooking Processes
Cooking Operation
G as B roiler H am burger
G as Fry er
G as G riddle
0%
20%
40%
60%
80%
100%
Particulate Removal Efficiency (%) Top-R ated B af f le
G eneric B af f le
H alton K S A
Figure 8. Particulate Grease Removal Efficiency for Three Filters With Different Cooking Processes
But what does the efficiency mean to and end-user? They are really concerned with how much grease will enter their ductwork, and deposit on the exhaust fan and roof of the facility. By taking the overall filter efficiency data and combining it with the amount of grease produced for a given cooking, the amount of grease which bypasses the filters can be calculated as shown in Figure 9.
Cooking Operation
G as B roiler H am burger
G as Fry er
G as G riddle
0
10
20
30
Grease Extracted by Filters (lb/1000 lb. food cooked)
Top-R ated B af f le
G eneric B af f le
H alton K S A
Figure 9. Mass of Grease Extracted by Filters With Different Cooking Processes If we look at the results from Figure 9, we can see that if an end-user chose a generic baffle filter instead of the Halton KSA filter, they would have approximately 14.5 more pounds of grease in the exhaust duct, on the exhaust fan and on the roof of the facility for every 1000 pounds of burgers cooked. Therefore, filter extraction efficiency plays a key role in fire safety and maintenance of a restaurant facility.
Click here for a full copy of ASTM F2519 Standard.
Return to the top of the page
Grease Extraction - Myth and Reality Grease Extraction Efficiency Test Protocol (VDI Standards) Click here to return to the Grease Extraction Page
Due to the expense, labor intensiveness, length of time and uncertainties associated with the EPA Method 5 approach, it was desirable to develop a new method of test (MOT) for determining the grease removal efficiency of devices. In Germany, VDI Standard 2052 Part 1 (VDI, 1999) was developed to specifically measure the particulate removal efficiency of kitchen filters as a function of particulate size. In the U.S.A. an ad-hoc group of manufacturers agreed in 2004 that this was the best type of approach for an MOT. In this approach, real-time data on grease emissions can be collected accurately as a function of particle size using an optical particle counter or size. This eliminates much of the labor cost, time lag and uncertainty associated with EPA Method 5 approach. The second large source of error involved with the EPA Method 5 is that it has used cooking to produce grease particles and contributes significantly to the uncertainties in the analysis. With the VDI method, a particle generator can be used to generate particles which improves the repeatability of particle generation and ultimately eliminates most of the uncertainties from cooking present in EPA Method 5. With the VDI Standard, particle size distributions before and after the tested filter are measured and used to calculate the fractional efficiency, i.e. the efficiency as a function of particle size. For each particle size (dp), the fractional efficiency Ef(dp) can then be determined as follows:
cb (db) Ef (dp) = 100 1- _________ ca (dp)
(
)
Where Ca (dp) and Cb (dp) refer to the upstream and downstream particle concentrations, respectively. Return to the top of the page
Grease Extraction - Myth and Reality Cyclonic Grease Extraction Click here to return to the Grease Extraction Page
One non-cloggable design of a baffle type grease extractor is a “cyclone.’ The extractor is constructed of multiple cyclones that remove grease from the air stream with the aid of centrifugal force. Figure 20 presents Halton’s KSA grease filter design. You can see the cyclonic action inside the KSA filter.
Figure 20 Halton KSA Filter
Figure 21 presents the extraction efficiency curve for Halton’s KSA filter for two different pressure drops across the filter.
Figure 21 Grease Extraction Efficiency Curves for KSA Filter
Mechanical grease filters quickly lose grease removal effectiveness as the particulate size drops below 5 to 6 microns depending on the pressure drop across the filters. As can be seen in Figure 21, the grease removal efficiency of mechanical filters when encountering PM 2.5 is 10% or less.
Return to the top of the page
Grease Extraction - Myth and Reality Ultra Violet Light Technology Click here to return to the Grease Extraction Page
Ultraviolet (UV) light had been used for years in water purification processes to destroy bacteria and other harmful microorganisms. In commercial kitchens, UV light has been applied to reduce or eliminate the grease emissions in the exhaust ductwork, on the exhaust fan, and on the roof of the building. It is important to realize that UV light technology is NOT a grease extractor; rather it is a grease conversion device. This distinction is necessary to understand since if a UV system were tested with the VDI 2052 Standard used with mechanical filters, an efficiency curve would not be generated. This section provides an overview of what the UV technology does and how it interacts with the mechanical extraction system (The Consultant Magazine, 2003). A general overview of how Halton’s UV technology is implemented is shown in Figure 10.
Figure 10. UV Grease Conversion Technology
UV Chemical Reactions There are two primary chemical reactions that take place in the UV oxidation process. The UV lights emit radiation in the UV-C band and also create ozone in the vicinity immediately surrounding the lamps. The chemical process taking place when UV-C directly hits molecular chains and breaks them into smaller compounds is called Photolysis. The photolysis reaction takes place in the immediate vicinity of the UV lamps (e.g., wherever light can hit the grease). The photolysis reaction is most effective on small grease particles (especially vapor) since the light can only break the chemical bonds on the outer surface of large grease particles. This is why it is critical to have efficient mechanical filtration prior to the UV lamps as seen in Figure 10. The second chemical process that takes place is when the ozone, created from the interaction of the UV light with the oxygen molecules in the air, continues to react with the grease molecules as they move through the exhaust duct to atmosphere. This process is called Ozonolysis. This ozonolysis reaction continues until either all of the ozone has reacted or exits the exhaust ductwork. After testing of a UV system while cooking at Halton Company, the plenum was disassembled and inspected. In the exhaust duct directly after the UV plenum, material collected on the walls of the duct. This substance was not greasy in feel but was powdery (see Photo 1). The smears in the photo are where the duct was wiped with a finger – which readily removed the powder from the duct surface. At the end of the 6 foot long duct transition the duct walls were reasonably clean (represented in picture provided). This provides an indication that the ozonolysis chemical reaction continues to work downstream in the ductwork and will clean the ductwork, exhaust fans and keep grease from building up on the roof of the building.
Photo 1. Powdery Substance Deposited on Ductwork after the Capture Ray™ System
Photo 2. Duct Is Clean Downstream of Capture Ray™ System
During testing it was also determined that the addition of ozone does not eliminate the odors present in the exhaust air, rather the odors are changed. While the odors may be minimized or eliminated for some light-duty cooking applications, there are no guarantees that odors will be eliminated. Therefore, in cases where odor removal is an important consideration, Halton recommends using the EcoloAirTM system downstream of the Capture Ray™ system. As previously stated in the chapter on Grease Defined, Myth vs. Reality, It is necessary to appreciate the nature of grease in commercial food service operations, the limitations of mechanical grease extraction, the effect of UV-C light on grease and the impact of temperature on the entire process. The Nature of Grease: An ASHRAE (American Society of Heating Refrigeration and Air Conditioning Engineers) research project, RP-745 documented emissions from different cooking processes. The report shows that the amount and size of grease particulate and vapor is function of the cooking process and product being cooked. The test method used was EPA Method 5. This test allows for the cooking of 1000 lbs. of product and measures total emissions. It further segments the emissions into 4 different classifications by total weight per 1000 lbs of cooked product, vapor content, weight of particles of 2.5 microns and smaller, weight of particles greater than 2.5 microns and less than 10 microns and weight of particles greater than 10 microns.
The Limitations of Mechanical Grease extraction: Once total emissions are estimated it is important to learn the limitations of mechanical extraction. Mechanical extraction is the primary means of removing larger grease particles in a UV system. These extractors should be tested in accordance with ASTM 2529 (Standard for testing of fractional efficiency based on 1000 lbs. of cooked product). Efficiency of mechanical extraction decreases the smaller the grease particle and conversely increases the larger the grease particle. This is due to the grease particles mass and the ability of the mechanical extractor to remove it with inertial force. Inertia is the primary energy used to “spin” grease out similar to slinging a wet towel. These results for the smaller particles and vapor are typical for mechanical extraction. This extractor is viewed as a high efficiency extractor since it is efficient above 90 percentile range on particles greater than 10 microns. Effect of UV on Grease: UV-C light in the range of 200 nanpmeters and below is the most effective for the break down of organic compounds. The effectiveness is based on loading (volume), particle size and resident time (exposure of grease to light). Effect of Temperature on UV-C lamp output The following curve shows the UV output of the lamps versus the surface temperature of the lamp. This shows the UV output at maximum for 42° C (108°F) and the ozone output would follow the general UV output curve. What has been observed in Halton’s laboratory testing is that we are working well when operating UV in the range of 150° to 160° F for exhaust temperature. Average exhaust temperature of a commercial kitchen exhaust system is in the 100° to 120° F range. Output for the lamps drops off at temperatures above 160° F.
Care should be taken when applying UV-C systems to high heat applications as well as loading on the lamps of grease particles and vapor. Typical application where this should be addressed is high volume under fired char broilers and high volume wok cooking.
Return to the top of the page
Click here to take a Halton quiz on Grease Extraction
