Collection Efficiency of Filters versus Impactors for airborn fungi

From Wikisource
Jump to navigation Jump to search
Warning: This page does not provide {{Header}} information. If you'd like to help, add a comment, or edit the page and replace "{{no header}}" with the following and fill in at least the title and author fields:
 | title      = 
 | author     = 
 | translator = 
 | section    = 
 | previous   = 
 | next       = 
 | year       = 
 | notes      = 
 | categories = 
 | portal     = 
Click here for a list of Author pages that link to this page.

This page is being considered for deletion at Proposed deletions. Please see the discussion there. Start discussion
Note for patrollers: if this page matches one of the criteria for speedy deletion, use {{sdelete}} instead.



Particulate matter in the air can lead to severe health problems. For example, bacteria and fungi, due to their ability to produce toxins are known to be a causal links in a number of infectious diseases. Fungi produce over 200 toxins. The best known toxins are the aflatoxins (potent carcinogens) produced by Aspergillus flavus and Aspergillus parasiticus, as well as, the satratoxins including Stachybotrys and Aspergillus versicolor that at high doses can cause immune system damage, affect the appetite center of the brain, and alter neurotransmitters.1 Low chronic doses commonly cause allergic rhinitis, hypersensitivity pneumonitis, asthma, and interstitial lung disease. It is estimated that $2.1 billion dollars are spent annually on consultant services for indoor air quality (IAQ) problem investigation, diagnosis, and resolution, in addition to the $500 million spent on IAQ litigation and insurance. Based on the available data, the cost of mold analysis alone is $50 million per year. Due to the health concerns and money spent annually on IAQ, it is relevant and justifiable to find a scientific sampling method that is capable of accurately and consistently quantifying fungal contamination of indoor air. Currently, a variety of methods are available for sampling ambient air. The two most prevalent methods, specific to indoor particulates, include glass slide impaction and filtration. The glass slide impactors are also known as spore traps. The Air-O-Cell® spore trap, a glass slide impactor, is designed to capture viable and non-viable particles from the air by drawing air through a rectangular slot at a recommended flow rate of 15 liters per minute (LPM) for a specified time of 5 to 10 minutes. The Allergenco D is a spore trap that draws air through an oval slot at a recommended flow rate of 5-20 LPM for durations of 1-20 minutes. The spores are impacted on a sticky medium that is spread over a clear plastic rectangular glass plate and then viewed microscopically for classification of type and enumerated for concentration. Collection efficiency of fungal spores by impactors must consider several parameters such as individual spore surface characteristics, initial percentage of aggregation of the spores, deaggregation rate during impaction, and the particle bounce rate. Tests of the Air-O-Cell® spore trap gave 100% collection efficiency for 3.2 μm particles at 15 LPM.3 This collection efficiency may not always be sufficient for collecting significant numbers of fungal spores smaller than this 100% collection size of 3.2 μm. The Air-O-Cell® spore trap has a recognized sample loss for particles less than 3 µm. The spore traps, Air-O-Cell® and Allergenco D, are also limited by the length of time they can sample. Collection of bioaerosols on filter media for microscopic analysis is a method utilized by industrial hygienists for several years and is referenced as an acceptable method of air sampling in the NIOSH Manual of Analytical Methods. Filters are typically housed in two or three piece, single-use plastic cassettes for field investigations. The plastic cassettes are designed to hold the filter in place and pressed together to provide a leak free connection between the two or three cassette pieces. Many cassettes are capable of being leak free up to 24 inches of mercury vacuum. Filter sizes of 25, 37, and 47 mm diameters are common for particulate sampling. The overall collection efficiency of a membrane filter is approximately 100% for particles larger than the pore size of the filter. However, filters are less commonly used due to the cost of enumeration. It would typically take a mycologist up to 5 hours to completely enumerate 20% of a 25-47 mm diameter trace. The following experiment tests the collection efficiency of a modified filter cassette, which was tapered to produce a reduced trace size, and compared to the standard impactors on the market today. It is hypothesized that the new design of filter cassettes will make them easier to enumerate, less costly to the consumer, and more efficient to test spore concentrations. The first type of filter cassette tested was the Bi-Air® filter cassette which houses a 25 mm mixed cellulose ester (MCE) filter and backer pad. The Bi-Air® cassette restricts the deposition area of the filter to two identical rectangular traces by pulling the air sample through a slotted-disk. The restriction is necessary because the complete area of a 25 mm filter is too large to economically analyze by microscopy. The second filter cassette tested was the Relle Smart Cassette prototype. The Relle Smart Cassette is a three-piece cassette designed to hold a 37 mm cellulose backer pad and MCE filter. The filter and backer pad are held in place by two nylon sections of the cassette, which are designed to restrict the air sampled through one rectangular trace. The third section of the cassette was an inlet section taken from a standard SKC, Inc. 37 mm filter cassette. Three trace sizes were used; a 2 x 20 mm, a 2 x 15 mm, and a 2 x 10 mm. All the Relle Smart Cassettes were loaded with SKC, Inc. 37 mm cellulose backer pads and SKC, Inc. 0.8 µm pore size MCE filters. Membrane filters are specified in Table II, Section F, of the NIOSH Manual of Analytical Methods as being good for very low to very high concentrations of bioaerosols. In addition, filter cassettes can sample for much longer periods of time resulting in a better representation of concentration over time which is important due to the viable microorganisms being metabolically active and given a stimuli can spontaneously release spores that contain toxins.

   The absence of a validated bioaerosol sampling method capable of accurately and consistently detecting fungal contamination in buildings, and the recognition that most methods underestimate actual concentrations in ambient air, investigators find interpretation of bioaerosol sampling results to be difficult.    Decreased collection efficiency can lead to a lack of identifying the different species present in the indoor air environment and due to the importance of concentration and type, it is necessary to establish, standardize and provide a reliable scientific method of detection and quantification.   Testing the accuracy and consistency of filters versus impaction is the first component, of many, that will lead to a reliable scientific methodology to increase the efficacy of investigating indoor air quality.


A settling chamber was constructed to provide a contained environment for testing and comparing the capture efficiencies of the Air-O-Cell® spore trap, Allergenco D® spore trap, Bi-Air® filter cassette, and Relle Smart Prototype filter cassette. Each device was placed in the settling chamber and exposed to aerosolized Aspergillus glaucans spores at a continuous rate and allowed to capture the spores for a predetermined length of time. After successful completion of a sampling period, the collection media were prepared for microscopic enumeration. The data was recorded and compared statistically within a 99% confidence interval.

Settling Chamber[edit]

The settling chamber was constructed of 2 cm plywood white finished on both the inside and outside. It was in the shape of a hexagon with 61 cm x 122 cm walls and provided a volume of approximately 1,400 liters. An access door was placed on one side as seen in Figure 1. Exterior View of Settling Chamber


Figure 1 Six sample ports, as seen in Figure 2, were place in the bottom of the settling chamber measuring 1 cm in diameter and 30.5 cm from wall of chamber and equally spaced. Six, 29 cm length by 4 mm outer diameter, steel tubes were inserted into the holes and used as the sampling ports. The six ports inside the chamber were numbered 1 through 6, with port #1 being located nearest the access door.

Interior View of Settling Chamber Example.jpg Figure 2

Each inside sampling port was connected to a brass fitting in the floor of the chamber and additional steel tubes were attached to each brass fitting under the chamber. These were routed underneath and to the side of the settling chamber for quick set up and attachment of pumps; see Figure 3. The steel tubes were then connected to six 4 ft long pieces of tygon tubing that connected the outside port to the appropriate pump.

Exterior (Sample tubing) Connector Ports Example.jpg Figure 3

The chamber make-up air was provided by using a seventh port located in the center of the chamber floor, as seen in Figure 2, which prevented excessive depressurization of the chamber.  A filtered SKC cassette was attached to the make-up air port to avoid contaminating the surrounding air or introduce contamination inside the chamber. 

The exhaust port, a 13 cm PVC pipe with an inner diameter of 2.5 cm, was located at the bottom of the chamber opposite the access door. The exhaust valve could be positioned as open or closed with the use of a toggle valve installed 8 cm outside the chamber. The exhaust pipe was attached to a HEPA filter industrial vacuum and used to evacuate the chamber at the conclusion of each experiment. Exhaust Port Connected to Hepa Filter Example.jpg Figure 4 The injection port, a PVC pipe with an inner diameter of 2.5 cm, was centered on top of the chamber. It could be positioned as open or closed with the use of a toggle valve installed 5.5 cm above the top of the chamber. Exterior Injection Port Example.jpg Figure 5 Attached to the injection port was a 1 cm outer diameter brass tube that was connected to the top of a nebulizer cup in train with a nebulizer. The nebulizer was utilized to aerosolize the fungal spores prior to entering the settling chamber for collection by the sampling devices. Attached to the side of the nebulizer cup, at a 90 angle was an adaptor and tygon tubing. The tubing connected a New Era Pump Systems, Inc., Multi-Phaser Model NE-500 Programmable Syringe Pump. The syringe pump was fitted with a Monoject® 60 cc Syringe with a catheter tip. The syringe pump was utilized as the fungal spore reservoir and programmed to inject the fungal spore mixture, 25% Aspergillus glaucans spores and Lab Safety Supply deionized water, at a rate of 2.5 ml per hour into the nebulizer cup for aerosolization prior to introduction to the interior of chamber. Nebulizer and Syringe Pump Example.jpg Figure 6

Spore Mixture[edit]

Aspergillus glaucans mold spores were isolated, grown, and prepared for injection in a clean room. The spores were grown in several Petri dishes containing potato dextrose agar and incubated at 25° C in a Quincy Lab, Inc., Model 12-140E Incubator. The surface of the colonies was washed into a beaker with a 50% solution of ACS grade isopropanol and deionized water. The ispropanol was used to make the mold spores go into solution with water. Aspergillus glaucans spores are hydrophobic and quickly separate from water if not exposed to a solvent. The beaker of 50% isopropanol and deionized water was left open over night in a Scienceware portable desktop hood to allow the isopropanol to evaporate.

Failure to remove the isopropanol from the liquid mixture would cause the MCE filters to react and prevent the flow of air through the filter.  The container was topped off with deionized water.  The water/spore mixture was stirred for one hour with a ceramic magnetic stirrer. The mixture was drawn into a 60 cc syringe and attached to the syringe pump for injection into the nebulizer.  


The pump assembly for the filter cassettes consisted of an internal Hargraves Technology Corporation model C175-12 vacuum pump, an Aalborg model GFM17 mass flow meter, battery and a small computer board.    The Aalborg mass gas flow meter is accurate to ±1.5% of full scale at temperatures between 32° and 122° F.12 The pump assembly communicated with an external data writing/processing board located near the pump via serial bus cable.  The pump assembly processing board read the information from a programmed data chip that directed the pump to sample at 1 LPM. Each filter cassette was paired with a Relle Smart Sampler prototype pump system for calibration using a primary standard.   

Interior Pump Assembly Example.jpg Figure 9

A DryCal DC-Lite Model DCL-H-Rev1.08 dry gas meter was used to calibrate each pump and cassette combination prior to sampling. The DryCal DC-Lite is accurate to 1.00%. The DryCal DC-Lite calibrator accuracy was verified monthly with a 500 ml graduated cylinder, soap bubbles and a stopwatch. After setting up the cassette and pump assembly, calibrating the cassette and pump assembly, the microchip for the pumps was programmed to sample at a predetermined rate and duration. A Gast model 1531-107-G557X high flow vacuum pump was used to pull the air through the spore traps. The rotometer accuracy was verified with a 500 ml graduated cylinder, soap bubbles and a stopwatch.

Filter Slide Preparation[edit]

The cassettes were separated at the A-B joint using a SKC Inc., 37 mm crow bar. The filter was removed from the cassette using a dental pick, held outside the trace area and cut with a pair of scissors to remove the excess filter. Using forceps the trace was placed face down on a 22 mm x 40 mm Hardy Diagnostics cover glass. The cover glass was placed on an Environmental Monitoring Systems, Inc., QuickFix Model 2122A Acetone Vaporizer slide tray. Trimming Filter Trace Placing Filter Trace on Slide Example.jpg Example.jpg Figure 10 Figure 11

The slide and tray was then inserted into the Acetone Vaporizer and ACS grade acetone was injected into the top of the vaporizer in 0.1 ml increments until the filter cleared and became transparent.  

Lactophenol Cotton Blue stain was placed in the center of a labeled microscope slide. The cover glass, with the affixed sample trace, was placed on the stain with the sample trace directly between the cover glass and microscope slide. The sample was allowed to sit for approximately one minute, then the cover glass was gently pressed down to thin the staining media and remove any bubbles from beneath the cover glass. The perimeter of the cover glass was painted with nail polish to keep the sample in place and the stain media from drying until microscopic counting could take place.

Spore Trap Slide Preparation[edit]

Lactophenol Cotton Blue stain was placed in the center of a labeled microscope slide. The spore trap was separated into two parts by cutting the paper label with a razor knife and prying the two parts with a SKC Inc., 37 mm crow bar. The two parts were separated and the plate with the sticky capture media removed. One drop of Lactophenol Cotton Blue was place on a cover glass and the sample placed face down on the center of the stain. The cover glass and sample was placed on a microscope slide with the sample facing up and allowed to sit for approximately one minute, then the cover glass was gently pressed down to thin the staining media and remove any air bubbles from beneath the cover glass. The perimeter of the sample was painted with nail polish for sample integrity.

Spore Counting Protocol[edit]

The microscope protocol and counting procedure was identical for both the filter and spore trap slides. Two Motic BA 300 microscopes were used for analysis. Both microscopes could examine at 600X power and were fitted with phase contrast and bright field objectives. One microscope included a camera attachment so images could be saved to a desktop computer. Motic BA 300 Microscope Example.jpg Figure 12 The slide, filter or spore trap, was placed on the microscope stage. The microscope was focused and the stage moved to one end of the trace. The beginning stage position was recorded on a data sheet and moved until the opposite end of the trace was observed. The ending stage position was recorded on the data sheet. The mycologist would start on one end of the trace, making passes across the short transit distance, and count the Aspergillus glaucans spores. The mycologist would continue transits until reaching a 500-spore count or upon completing the final transit. The spore count after completing the final transit was recorded on the data sheet and the ending stage position was recorded on the data sheet. The start and stop position ratio compared to the start and opposite end positions was determined. This ratio was multiplied by the remaining transits to ascertain an estimated total number of spores collected on the slide. If the slide did not have more than 500 Aspergillus glaucans spores, the entire trace area was counted.



Model R R Square Adjusted R Square Std. Error of the Estimate 1 .550(a) .303 .264 .84359 a Predictors: (Constant), Filters b Dependent Variable: Air-O-Cell®

Correlation between the Filters and Air-O-Cell®: R square = 0.303: Interpretation: 30.3% of variability in Y (Air-O-Cell®) is explained by the linear relationship between the Air-O-Cell® and the Filters.


Model Sum of Squares df Mean Square F Sig. 1 Regression 5.564 1 5.564 7.819 .012(a)

   Residual        12.810  18      .712             
        Total   18.374  19                       

a Predictors: (Constant), Filters b Dependent Variable: Air-O-Cell®

Analysis of Variance: Ho = No relationship between Y and X

                      Ha = Linear relationship between Y and X

Test Statistic: F = 7.819, df (1, 18) P-value = 0.012

Decision: We reject Ho in favor of Ha and conclude the data has demonstrated a linear relationship between the Air-O-Cell® and the Filters. (F = 7.89, df (1, 18), p= 0.012)

T = 2.796, p = 0.012 we reject Ho in favor of Ha and conclude the Air-O-Cells and Filters are linear.

Correlation between the Filter Prototype and Impactor (Allergenco D®): R square = 0.921: Interpretation: 92.1% of variability in Y (Allergenco D®) is explained by the linear relationship between the Allergenco D® and the Filter Prototype.

Analysis of Variance: Ho = No relationship between Y and X

                      Ha = Linear relationship between Y and X

Test Statistic: F = 61.679, df (1, 11) P-value = 0.00

Decision: We reject Ho in favor of Ha and conclude the data has demonstrated a linear relationship between the Allergenco D® and the Filters. (F = 61.679, df (1, 11), p= 0.00)

T = 7.854, p = 0.00 we reject Ho in favor of Ha and conclude the Allergenco D® and Filter Prototype are linear.


A total of twenty-three experiments were taken in the settling chamber, approximately 90 samples. All experiments were injected with 4-5 µm Aspergillus glaucans fungal spores. The samples were taken over two-hour periods with the spores being injected into the chamber throughout the sampling period. Each sample trace was analyzed, enumerated, and recorded. Comparison of the collection efficiencies using 4-5 µm Aspergillus glaucans spores in a settling chamber demonstrated a better collection efficiency of 1.6 to 1.7 times for the overall Prototype and Bi-Air® membrane filter media as compared to Air-O-Cell® spore traps. Additionally, the comparison of the efficiencies using 4-5 µm Aspergillus glaucans spores demonstrated a collection efficiency of 2.2 times for the Prototype membrane filter media as compared to Allergenco D® spore traps. This data supports the J. Spurgeon data in which the collection efficiency of the Bi-Air® compared to the Air-O-Cell® for Penicillum chrysogenum spores was demonstrated to be 1.7 to 2.0 times greater for the Bi-Air® membrane filter cassette. 9


The experiment concludes within a 99% confidence interval that the collection efficiency of a membrane filter versus an impactor spore trap is linear and establishes that the collection efficiency in a controlled environment is better for filters. Due to the biologic versatility of fungal spores and the fact that type and concentration are so relevant to health, litigation, and money spent annually for investigation and remediation; it is recommended that sampling for indoor airborne fungal matter be conducted on filter media vice spore traps. A more accurate total concentration and potential exposure can be assumed by the longer sampling periods of filters. Further research should be conducted to evaluate filters versus spore traps in non-controlled environments such as contaminated buildings and heating and ventilation systems. Limitations of using the filter media include uneven filter deposition and occlusion in heavy concentrations; however, further design and structural research of the collection cassette can corrected this deficiency. For example, incorporating a dispersion nozzle in the airflow and increasing the length of mixing chamber could even the filter deposition and decrease the amount of time and money spent on laboratory analysis while still acquiring a more accurate count and type of fungal contamination. TulaneGrad 04:19, 24 July 2007 (UTC)TulaneGradTulaneGrad 04:19, 24 July 2007 (UTC)


  • Ljungqvist, Bengt. 1998. Active Sampling of Airborne Viable Particles In Controlled Environments: a Comparative Study of Common Instruments. Eur J Parenteral Sci. 3:59-62.