United States
Environmental Protection
Agency
-------
-------
EPA/600/R-08/032 November 2008 www.epa.gov/ord
FINAL REPORT ON
Assessment of Advanced Building
Air Filtration Systems
Contract No. GS-10F-0275K
Task Order 1105
Prepared for
Joseph Wood and Les Sparks, Project Officers
U.S. ENVIRONMENTAL PROTECTION AGENCY
Research Triangle Park, NC 27511
Prepared by
Anbo Wang, Ph.D., and Kent C. Hofacre
"WARNING - This document may contain technical data
whose export is restricted by U.S. law. Violators of export
control laws may be subject to severe legal penalties. Do
not disseminate this document outside the United States
or disclose its contents to non-U.S. persons except in
accordance with applicable laws and regulations and after
obtaining any required authorizations."
BATTELLE COLUMBUS OPERATIONS
505 King Avenue
Columbus, Ohio 43201-2693
Office of Research and Development
National Homeland Security Research Center, Decontamination and Consequence Management Division
-------
Disclaimer
This report is a work prepared for the United States Government by Battelle. In no event shall either
the United States Government or Battelle have any responsibility or liability for any consequences
of any use, misuse, inability to use, or reliance on the information contained herein, nor does either
warrant or otherwise represent in any way the accuracy, adequacy, efficacy, or applicability of the
contents hereof.
-------
Table of Contents
Page
Executive Summary viii
1.0 Introduction 3
2.0 Objective 3
3.0 Scope 3
4.0 Literature Review and Performance Assessment 5
4.1 Literature Search 5
4.2 Establishment of Performance Requirements 6
4.3 Candidate Technologies Assessment 7
4.3.1 Electret Media 7
4.3.2 Electrically Enhanced Filters 10
4.3.3 NanofiberFilter 12
4.4 Summary of Literature Review 13
5.0 Candidate Media Evaluation Tests 15
5.1 Test Methods and Procedures 15
5.1.1 Airflow Resistance 15
5.1.2 Aerosol Collection Efficiency 16
5.1.3 Laboratory Conditioning and Efficiency Evaluation 18
5.1.4 Ambient Conditioning and Efficiency Evaluation 18
5.2 Test Results and Discussions 19
5.2.1 Airflow Resistance 19
5.2.2 Initial Penetration Fraction 20
5.2.3 Quality Factor 24
5.2.4 Laboratory Conditioning 24
5.2.5 Ambient Conditioning 29
5.3 Summary of Experimental Study 30
5.4 Advanced Filter Development 30
6.0 Conclusions and Recommendations 31
7.0 References 33
Appendix A Descriptions of the Databases Searched A-l
-------
List of Figures
Page
Figure 1. Schematic of the Airflow Resistance Test System 15
Figure 2. Schematic of the Aerosol Efficiency Test Setup 17
Figure 3. Schematic of the Ambient ConditioningTest Setup 18
Figure 4. The Initial Airflow Resistance of Candidate Electret Media 20
Figure 5. The Initial Airflow Resistance of Candidate Nanofiber Media 20
Figure 7. Initial Penetration Fraction of Nanofiber Candidate Media 21
Figure 6. Initial Penetration Fraction of Electret Candidate Media 21
Figure 8. Initial Collection Efficiency of Candidate Media at Varying Velocities 22
Figure 9. Size Distribution of the Conditioning Aerosol (Measured by SMPS) 25
Figure 10. Incremental Laboratory Conditioning of Sample A Electret Media 25
Figure 11. Incremental Laboratory Conditioning of Sample F Electret Media 26
Figure 12. Incremental Laboratory Conditioning of Sample G Electret Media 26
Figure 13. Incremental Laboratory Conditioning of Sample G Electret Media 27
Figure 14. Penetration Fraction as a Function of Conditioning Time (for Sample G Electret Media) 28
Figure 15. The Airflow Resistance of Candidate Electret Before and After Conditioning 28
Figure 16. Size Distribution of the Indoor Ambient Conditioning Aerosol (Measured by WPS) 29
Figure 17. Indoor Ambient Conditioning of Sample G Electret Media 30
List of Tables
Page
Table 1. Literature Search Strategies and Summary Results 5
Table 2. Leading High-Efficiency Filters Available in the Market 6
Table 3. Performance Requirements of the Advanced Filtration System 7
Table 4. List of Electret Filter and Media Manufacturers 9
Table 5. Performance Data of Candidate Electret Media from Major Manufacturers 11
Table 6. Performance Data of Electrically Enhanced Filter Systems 11
Table 7. Technical Properties of Donaldson Nanoweb Media 12
TableS. Test Matrix 19
Table 9. Quality Factor Comparison of the Candidate Sample Media 24
-------
List of Acronyms
ASHRAE American Society of Heating, Refrigerating and Air-Conditioning Engineers
CB chemical/biological
CBIAC Chemical and Biological Information Analysis Center
CMD count mean diameter
DTIC Defense Technical Information Center
EEF electrically enhanced filter
EPA Environmental Protection Agency
HEPA high-efficiency paniculate air
HVAC heating, ventilating, and air conditioning
KC1 potassium chloride
MERV Minimum Efficiency Reporting Value
NTIS National Technical Information Service
ORD Office of Research and Development
QF quality factors
SMPS Scanning Mobility Particle Sizer
TOPO Task Order Project Officer
-------
Executive Summary
The original purpose of the work described in this report
was to develop an advanced air filtration system that could
be adapted to a building's HVAC system to help remove
biological agents from the building environment. It was
desired that the advanced filtration system provide lower
pressure drop than conventional high-efficiency paniculate
filters with higher or equivalent efficiency and comparable
or lower cost. Through literature searches, market surveys,
technology assessments, and discussions with air filtration
system manufacturers, it was determined that new technology
was not adequately advanced to merit development of an
advanced HVAC paniculate removal system for this project.
Improvements in technologies were identified, but nothing
that could improve performance much beyond what existed
in current or soon to be commercially available products.
Therefore, in lieu of describing the development of an
advanced filtration system, this report provides an assessment
and discussion of advanced particle removal technologies for
HVAC systems.
As the first step of this study, the performance requirements
of an advanced filtration system were established to provide
a basis for evaluation of candidate technologies. The
requirements were established considering two criteria: (a)
has better performance than the high efficiency filters (MERV
14, 15, and 16 filters) available in the market and (b) does not
exceed the pressure drop limit that common HVAC systems
can accommodate. Based on these criteria, the performance
requirements established were a 99.9% removal efficiency
for aerosols with a 1-um diameter (optical diameter) and with
a pressure drop of less than 0.5 in. H2O. It should be noted,
however, that there are no defined "safe" levels for biological
agents, thus this performance requirement cannot ensure that
exposure to an infectious concentration of a biological agent
will be prevented.
A comprehensive literature search was conducted to identify
advanced filtration technologies and manufacturers that
could potentially be used in the advanced filtration system.
The databases searched, including CBIAC, DTIC, CA
Search, NTIS, Energy SciTec, Ei Compendex®, SciSearch®,
and Biosis Previews®, covered a wide variety of technical
journals, conference proceedings, patents, government
reports, and books. A market survey was conducted
simultaneously with the literature search through the Internet,
phone conversations, and meetings with leading filter/filter-
media manufacturers at professional conferences.
Three filtration technologies were identified as preliminary
candidates for an "advanced" system: electret filters,
electrically enhanced filters (EEFs), and nanofiber media
filters. The operation principle, potential drawbacks,
technology maturity, ability to meet the performance
requirements, and cost of each candidate technology were
assessed. Upon further analysis, EEFs were rejected as an
advanced filtration technology because of their relatively
high cost compared to conventional filters, as well as their
potential diminished collection efficiency with dust loading.
Sample electret and nanofiber media were requested
from manufacturers. Screening tests were conducted
to measure the initial aerosol collection efficiency and
airflow resistance. As shown in Equation ES-1, the
performance of the candidate media was compared
to the performance requirements and ranked using a
systematic parameter called quality factor (QF):
QF =
APx6
(ES-l)
where: p is the penetration fraction of 1-um particles,
AP is the pressure drop (mmH2O), and
5 is the filter media thickness (mm).
Three candidate nanofiber media with different expected
collection efficiencies were tested but demonstrated QFs
significantly lower than the hypothetical advanced filter
QF. Therefore, the nanofiber technology was excluded
from further evaluation. Among the eight electret media
tested, three demonstrated QFs higher than the performance
requirements. In other words, those three media offered lower
penetration, lower resistance, and/or were thinner than a filter
just meeting the specification.
The literature reported the potential degradation of electret
filters with aerosol loading and the importance of identifying
the minimum collection efficiency of an electret filter
within its service life (Lehtimaki and Heinonen, 1994;
Lehtimaki et al., 1996; Lifshutz, 1997; Pierce and Lifshutz,
1997; Barrett, 1998; Hanley et al., 1999; Raynor and
Chae, 2002; Raynor and Chae, 2003). Thus, a laboratory
conditioning method was developed by Hanley and Owen
(2003) to try to identify the minimum collection efficiency
of an electret filter. The three electret sample media that
demonstrated promising QFs were conditioned in the
laboratory using a nano-sized KC1 aerosol, according to
the method developed by Hanley and Owen (2003). The
candidate electret samples were also conditioned with
ambient aerosol to characterize the degradation in an actual
ambient environment. Conditioning tests were performed
in incremental steps, with efficiency measured after each
increment, to identify the minimum collection efficiency.
Among the three sample media tested, only Sample G
demonstrated excellent efficiency stability. Both initial and
minimum collection efficiencies of Sample G (based on
laboratory conditioning) met the efficiency goal of 99.9%
for 1-um diameter particles. The initial airflow resistance,
however, was approximately 6% higher than the "advanced"
filter performance requirement. The slightly higher airflow
-------
resistance can be reduced by enlarging the filter media design
area to over 100 ft2 (i.e., the design area that the media
evaluation tests are based on), which is attainable since a
typical pleated high-efficiency HVAC filter usually has media
area ranging from 100 to 180 ft2.
In conclusion, tests conducted in this study with swatches of
candidate filter media demonstrated the potential to develop
an advanced electret filter that can meet the performance
goals. It was determined, however, that the incremental gain
in collection efficiency and reduction in airflow resistance
were not sufficient to merit continuing with the development
of the advanced filter under this project. Furthermore, the
manufacturer of the leading media (Sample G) indicated that
they had already planned further improvements to that media
and that a filter made of the improved media was being tested
and was expected to be on the market soon.
-------
-------
1.0
Introduction
Buildings can be vulnerable to terrorist attacks using
various types of threat agents. The most serious effects
of such an attack are on the health of the occupants of
the buildings. Building occupants may suffer health
effects ranging from irritation, to severe sickness, to
death. The attack may also have long-term economic and
other impacts due to building contamination. Although
guidelines exist, there is still some uncertainty as to the
optimum course of action to take in mitigating the impact
of a terrorist attack on a building. Tools and technologies
to implement optimum courses of action are often not
available, are too expensive to use, or are insufficient.
2.0
Objective
The original purpose of the project described in this report
was to develop an advanced air filtration technology that
could be adapted to a building's HVAC system to protect
a building from a biological attack. The advanced air filter
would provide a lower pressure drop than conventional
high-efficiency paniculate filters (MERV 14, 15, and
16 filters) but with higher or equivalent efficiency and
comparable or lower cost. But as explained in the
following sections, the focus of the project switched
from the development of an advanced technology to
an assessment of currently advanced technology.
3.0
Scope
The first step of the project was to conduct a literature
review and market survey to: (a) identify candidate
advanced air filtration technologies that could potentially
be used in protecting the indoor environment from
biological agents and (b) establish performance
requirements for the advanced filtration system to be
developed. The approaches and results of the literature
review are presented in Section 4.0 of this report.
Sample candidate filtration media (based on two filter
technologies) identified in the literature review were
requested from the corresponding manufacturers. The
sample media were evaluated experimentally to explore
the feasibility of developing an advanced filtration system
that can meet the performance goals. The test methods, data
collected, and the results of the evaluation tests are presented
in Section 5.0.
Based on all of the efforts of this project (literature searches,
market surveys, technology evaluations, and discussions
with air filtration system manufacturers), it was determined
that new technology was not adequately advanced to merit
development of an advanced paniculate removal system
under this project. Improvements in technologies were
identified, but nothing that could improve performance
much beyond what was already available or soon to be
commercially available. Nonetheless, an assessment and
discussion of advanced particle removal technologies for
building HVAC systems is provided.
-------
-------
4.0
Literature Review and Performance Assessment
4.1 Literature Search
A literature search was conducted to identify advanced
filtration technologies that could be developed for an
HVAC system. Eleven databases — CBIAC, DTIC, CA
Search, NTIS, Energy SciTec, Ei Compendex®, SciSearch®,
Biosis Previews®, Enviroline®, World Textiles, and Textile
Technology Digest — were searched through the Chemical
and Biological Information Analysis Center (CBIAC) and
the Dialog online information systems. These databases were
selected to ensure that the literature search covered a wide
variety of technical journals, conference proceedings, patents,
government reports, and books. Brief descriptions of the
eleven databases are presented in Appendix A.
Based on information obtained in related research regarding
air filtration, the literature search first focused on the
following technologies: (1) electret filtration media, (2)
nanofiber filtration media, (3) filters with biocides, and
(4) layered composite filters. A more general search was
also conducted to identify other potentially advanced
technologies in the air filtration area. The search strategies
and the corresponding hits generated are summarized in
Table 1. Note that the initial search in the area of electret
filtration media generated a large number (12,000) of hits.
Subsequently, the search was narrowed using the keywords
"review(s) or survey(s)."
Table 1. Literature Search Strategies and Summary Results
A total of 2,060 hits were generated using the search
strategies presented in Table 1. The titles and/
or abstracts of the 2,060 hits were screened, and 50
relevant articles were identified and ordered. Relevant
articles collected for other previous related research
were also reviewed. Note that no relevant articles were
identified for the layer composite technology, so this
was removed from the candidate list of technologies.
In addition, a market survey was conducted through the
Internet, phone conversations, and meetings with potential
leading manufacturers at professional conferences. A brief
Internet search was conducted to identify manufacturers
in the areas of electret media, nanofiber media, and any
other novel media that had high efficiency and relatively
low pressure drop. Battelle staff also met with sales
and technical representatives from the major filter and
filtration media manufacturers during the Filtration
2004 International Conference and Exposition on 7-9
December 2004, in Philadelphia. Follow-up phone
discussions were held with the manufacturers who carried
the products of interest to request further technical and
cost data for a preliminary screening and evaluation.
_ . . _ . _. . „ , .... # of Articles
Target Area Search Strategy # of Hits
Electret Media
Nanofiber Media
Biocidal Media
Layer Composite Media
Advanced and Novel
filters
Air Filter Review
{filter? or filtration or media} and {electret? or electrostatic?} and {review? or
survey?}
{filter? or filtration or media} and {nanofiber? or nanofibre?}
{filter? or filtration } and {biocidal}
{filter? or filtration } and {layer?(5N) composite?}
{air} and {filter?} and {advanced? Or novel or (state (2w) art)}
{air} and {filter?} and {review? Or survey?}
Total
374
256
100
251
474
605
2060
25
7
7
0
5
6
50
-------
4.2 Establishment of Performance Requirements
Currently, there are no performance criteria established for
HVAC air nitration systems designed to protect building
occupants against biological agents because there are no
denned "safe" levels of exposure to biological threat agents.
To help down-select technologies for consideration in an
advanced nitration system, performance requirements were
established not based on risk reduction but on: (a) the
collection efficiency of niters available in the market, (b)
the range of aerosol sizes expected to be representative of
bio-terrorist attacks, and (c) the maximum pressure drop that
common HVAC systems can accommodate.
A brief market survey was conducted to identify the best
niters available in the market for commercial HVAC
application. Although HEPA niters provide high nitration
efficiency, they are not necessarily appropriate for HVAC
applications. As a general rule, existing HVAC systems
cannot be upgraded to HEPA niters without a complete
retrofit of the air handling system due to the high pressure
drop and potential leakage associated with them. Instead,
filter manufacturers (AAF International, 2005; AIRGUARD,
2004) recommend high-efficiency filters (MERV 15 and 16)
as a cost-effective alternative to HEPA filters for maximum
paniculate removal. Table 2 lists a sample of the high-
efficiency filters available in the market as well as a summary
of their performance and cost. As shown in Table 2, the
leading high-efficiency filters in the market demonstrate
comparable performance. The filters with uncharged media
can provide efficiency ranging from 96 to 99% for 1-um
diameter particle and pressure drop ranging from 0.4 to 0.6
in. H2O, with a cost ranging from $170 to $230. Special
designs such as V-bank, mini-pleat, or V-shape pleat are
applied in these filters to reduce pressure drop.
The high-efficiency electret filters can provide a slightly
lower pressure drop (0.27 to 0.35 in. HO), but with an
equivalent initial efficiency (95 to 98% for 1 um) and
lower cost compared to the uncharged filters. The potential
degradation of the commercial electret filters, however,
remains a concern.
The performance of the high-end filters in the market
was used to establish the performance requirements of an
advanced filtration system. The priority of performance
criteria in developing the advanced filter, in decreasing order
of importance, is pressure drop, collection efficiency, cost,
and size.
The performance goal for collection efficiency was specified
for 1-um particles because most bio-aerosol challenges are
expected to have diameters equal to or larger than 1 um.
Furthermore, only aerosols with diameters ranging from 1
to 5 um can be transported long distances by wind without
decay and can be inhaled deeply into the lungs (Edward,
1997). The market survey revealed that the high-efficiency
filters currently commercially available could achieve a
collection efficiency near 99% for 1-um particles. For the
hypothetical advanced filter being considered in this study, it
was determined that the collection efficiency would have to
be an order of magnitude better, i.e., it would have to have an
efficiency higher than 99.9% for 1-um particles.
The performance goal regarding pressure drop was set to
be less than 0.5 in. H2O at a face velocity of 500 fpm to
ensure the advanced filter could be used in an existing HVAC
system without extensive modification (i.e., retrofitting with
a larger blower unit). Pressure drop across the mechanical
filters in a typical HVAC system in a standard office building
is generally less than or equal to 0.5 in. H2O. By setting the
performance goal to be less than 0.5 in H2O, the developed
filter could be installed into a standard office building HVAC
system without modifications. In comparison, if a HEPA
filter were installed into an existing HVAC system, major
modifications would need to be made since the pressure drop
of a HEPA filter typically ranges from 1 to 2 in. HO.
Table 2. Leading High-Efficiency Filters Available in the Market (Filter Size: 24" x 24" x 12")
Initial AP Cost($)
Media Type Manufacturer Model Filter Type TTfor l(xm (jn H Q) per filter
@500 fpm @50Q 1 (2 000 cfm)
HEPA
Uncharged
Charged
[b]
Microglass
paper fibers
Synthetic
media
Glass fiber
papers
Synthetic
media
Synthetic
media
[b]
AAF
AIRGUARD
Freudenberg
Freudenberg
TOYOBO
[b]
VariCel® V
VARI+PLUSeVP
Viledon® MX98
Viledon® MV95
SL-56-95T
Mini-pleat,
V-Bank
Mini-Pleat,
V-Bank
V-shape
pleats, box
Pleat,
V-Bank
Pleat, Box
>99.99
97%
> 96%
> 99%
> 98% (initial)
95%
1 to 2
0.59
0.4
0.46
0.35
0.27
180 to 500
170
231
Jc]
152
Jc]
[al T| is defined as collection efficiency.
[bl Representative of a typical HEPA filter is AstroCel HCX (HEPA) filter from AAF® International.
[cl The cost was not provided by the manufacturer.
-------
To ensure the comparison of technologies is made on
an equivalent basis, the required efficiency and airflow
resistance listed in Table 3 are based on an air handler
capacity of 2,000 cfm and filter dimensions not larger than
24" x 24" x 12".
Table 3. Performance Requirements of the Advanced
Filtration System
Filter Parameter Goal
Efficiency
Airflow Resistance
Filter Dimensions
> 99.9%
for l-|jm aerosol
(At 2, 000 cfm or 500 fpm)
<0.5 in. H20
(At 2,000 cfm or 500 fpm)
< 24" x 24" x 12"
4.3 Candidate Technologies Assessment
Based on the aforementioned literature review, market
survey, and discussions with sales/technical representatives
from major manufacturers of air filters, the following
technologies were identified as candidates for the advanced
filtration system: (1) electret filters, (2) electrically enhanced
filters (EEF), and (3) nanofiber media filters. No promising
filters were identified using biocidal additives or layered
composite technology. The candidate technologies are
described and assessed in the following sections.
4.3.1 Electret Media
4.3.1.1 Technology Description. Many air filters in the
market are currently manufactured using electrically charged
media to attract particles. This improves a filter's efficiency
without increasing its pressure drop. Filters that use this
technology are commonly referred to as "electrostatic,"
"electrically charged," or "electret" media. The advantage
of electret media is their relatively high collection efficiency
at relatively low pressure drops, when compared to filters
relying solely on mechanical means for particle capture.
Electret media are made of dielectric materials that have a
significant microscopic bipolar charge on the fibers and a
very low net macroscopic charge. Unlike the electrically
enhanced filters described in Section 4.3.2, electret media are
permanently charged during media manufacturing. Therefore,
electret media do not require an electrode system to charge
filter media or an ionizer to charge incoming particles
during operation. Electret filters collect particles through a
combination of conventional mechanical mechanisms (i.e.,
impaction, interception, and diffusion) and electrostatic
mechanisms (i.e., Coulombic attraction and dielectrophoretic
capture). Charged particles are attracted to oppositely charged
fibers by the Coulombic force. For singly charged particles,
the attraction increases as particle size decreases. Neutral
particles that are unaffected by Coulombic force are collected
by dielectrophoretic force—the polarization force induced
by local electrical fields within the filter media. Charged
particles are also collected by dielectrophoretic capture. The
efficiency of the dielectrophoretic capture increases with
particle size.
The efficiency of electret media depends on parameters
such as charges on particles, charge density of fibers, and
chemical compositions of particles and fibers; efficiency also
depends on factors that affect the efficiency of conventional
uncharged filters, such as fiber diameter and packing density
of the fibrous materials.
There are many types of electret media, due to the variety
of fiber-forming technologies (i.e., meltblown, split fiber,
bi-component spunbond, needlefelt) and the variety of
electrostatic treating technologies (i.e., corona charged,
triboelectric charged, induction charged). The composition
of electret media varies from polycarbonate, polypropylene,
and polyolefin to a binary mixture of polypropylene and
chlorinated acrylic fiber. Because the media are manufactured
using different technologies and are composed of different
polymers, there is a significant range in filtration performance
and degradation behavior (Barrett and Rousseau, 1998;
Romayetal., 1998).
4.3.1.2 Potential Drawbacks. A concern with using
electret filters is the effect of aerosol loading on collection
efficiency. The collection efficiency of an electret filter for
solid particles has been found to decrease with operation time
in its early stage of collection until it loses electrical forces.
At that point, the collection efficiency stabilizes but then
increases with time because the filter media become loaded
with the solid particles (Myers and Arnold, 2003).
Electret filters also degrade when loaded with oil aerosols
(Lehtimaki andHeinonen, 1994; Lifshutz, 1997; Pierce and
Lifshutz, 1997; Barrett and Rousseau, 1998). Oil-resistant
electret filters, which have much lower degradation by oil
aerosols, were developed and used in particle respirators
(Barrett and Rousseau, 1998; Romay and Liu, 1998;
Janssen et al., 2003a and 2003b). Because oil aerosols are
not the major components of ambient/indoor aerosols, the
assessment in electret degradation of this study focused on
the degradation by solid aerosols.
Arizona road dust is the ASHRAE test dust that is currently
used in the conditioning step of the ASHRAE Standard
52.2. Several studies (Lehtimaki, 1996; Hanley et al., 1999;
Raynor and Chae, 2002; Raynor and Chae, 2003) revealed,
however, that the degradation of the electret filter, when
loaded with the ASHRAE dust, is less significant than
when the filter was exposed to real ambient conditions.
These studies revealed that the ASHRAE 52.2 dust-loading
procedure does not adequately reproduce the reduction in
filtration efficiency that an electret filter encounters in actual
HVAC systems. The ASHRAE Standard 52.2, which was
developed to determine the minimum efficiencies of a filter
over its lifetime, may actually provide an artificially higher
MERV rating for an electret filter.
Realizing the potential deficiency of the ASHRAE Standard
52.2 that tends to show an artificially higher MERV rating
for electret filters, the ASHRAE committee supported a
research project conducted by Research Triangle Institute
(RTI) to develop a dust for a new loading test method that
-------
will more accurately determine the minimum efficiency
points of an electret filter in a real-world application (Hanley
and Owen, 2003). Under this project, a new loading test
method was developed to replace the first dust loading step
(or the conditioning step) of ASHRAE 52.2, using a nano-
sized solid-phase KC1 aerosol (with number mean diameter
of 0.035 um) as the conditioning aerosol. The new method
provided a means of accelerating the decrease in efficiency
that electret filters undergo in real-life applications. A draft
addendum (Addendum C) to ASHRAE Standard 52.2 was
prepared during the project. The addendum includes a
detailed protocol of conditioning the electret filters using
nano-sized KC1 aerosols to mask (or screen) the charges on
the electret filter. ASHRAE Standard 52.2 Addendum C is
currently available for public review.
Despite the potential efficiency degradation of electret
media with use, they have gained significant market
share and acceptance in HVAC filtration applications
over the past few years (Arnold and Myers, 2002;
Homonoff, 2004). This is because electret filters are
usually less expensive than mechanical filters (glass
fiber filters) with the same MERV rating. In addition,
in spite of the collection efficiency degradation, the
efficiency of an electret filter will always exceed that of an
uncharged filter with the identical mechanic structure.
When selecting an electret filter for an HVAC filtration
application, it is important to evaluate the electret filter
performance data at specific application conditions. If in-use
performance data are not available, a laboratory loading test
(which can represent the minimum efficiency points of an
electret filter in a real-world application) should be conducted
to ensure the selected electret filter can meet the design goals
of a particular HVAC application.
4.3.1.3 Technology Maturity for Use in the Advanced
Filter. In the HVAC filtration market, electret filters are
finding increased popularity (Myers and Arnold, 2003;
Homonoff, 2004). Nearly all high-efficiency (MERV 11 or
higher) residential filters are composed of electret material
as well.
The electret filters available for residential HVAC filtration
generally have MERV ratings ranging from 8 to 12. The
typical pressure drop for residential pleated electret filters
ranges from 0.13 to 0.35 in. H2O at 300 fpm (3M Brochure,
Improve Indoor Air). The electret filters used for commercial
HVAC filtration generally have MERV ratings ranging
from 8 to 16. The two major manufacturers of electret
filters are Freudenberg Nonwovens and the 3M Company.
Freudenberg, a leading manufacturer of commercial HVAC
electret filters, developed and patented a process in which
polymer fibers (e.g., polycarbonate fibers) are spun in an
electrostatic field. This process is known as the electrostatic
spinning process. Since the fibers are manufactured in an
electrostatic field, they carry an electrostatic charge, which
significantly improves the collection efficiency.
Freudenberg's pleated electret filter, Viledon® MV95 (MERV
15), as shown in Table 2, has over 98% efficiency for a
1-um aerosol and a pressure drop of only 0.35 in. H2O at
500 fpm. This performance of Viledon® MV95 is close to
the design goals presented in Table 3. According to David
Matier, Manager of North American Operations at the
Freudenberg Group, the Viledon® MV95 is currently used in
the HVAC systems of the federal buildings of Los Angeles
and Honolulu.
The high performance of the Viledon® electret filter makes
the Viledon® electret media one of the top candidates to be
considered for use in the advanced filtration system. Battelle
contacted and discussed with Dr. Andre Manz, a senior
applications engineer at Freudenburg, the development of
the advanced filtration system. According to Dr. Manz, the
design goals presented in Table 3 are challenging. However,
they may be achievable by improving the design of the
Viledon® MV95 filter by adding more filter media.
Freudenberg sent a sample of the electret material (4 ft2)
used in the Viledon® MV95 filter to Battelle for evaluation.
Samples of an MV95 filter and an MF95 filter were also
received from Freudenberg.
The 3M Company specializes in producing high-end electret
filters for residential HVAC application. 3M's residential
electret filters, Filtrete™ Ultra Allergen, Filtrete™ Micro
Allergen, and Filtrete™ Dust & Pollen filters are rated as
MERV 12, 11, and 8, respectively.
3M fabricates three types of electret: corona-charged
split-fiber media, corona-charged meltblown media, and
advanced electret media. The corona-charged split-fiber
media, with a commercial name of Filtrete™ Type G,
are made from fibrillation of a polypropylene thin film
charged by corona ions. The corona-charged meltblown
media, with commercial names of Filtrete™ Types B, E,
and S, are charged by corona ions during the meltblowing
process. The patented advanced electret media are a
new class of filter media developed by 3M and used in
3M's N95, P95, and P100 paniculate respirators.
Battelle contacted Dr. Michael Strommen, the product
development manager from 3M Filtration, to discuss
the development of the advanced filtration system.
According to Dr. Strommen, in addition to the well-
known Filtrete™ electret filters for residential HVAC
applications, 3M also fabricates high-efficiency electret
filters for commercial HVAC applications. The technical
data sheet of a 3M commercial high-performance
HVAC filter (MERV 14) was sent to Battelle. Similar
to Dr. Manz, Dr. Strommen also believed the design
goals were challenging but may be achieved using a
V-bank design to accommodate more filter media.
Samples of two grades of 3M meltblown electret media were
sent to Battelle in April 2005. According to Dr. Strommen,
the two grades are at the high end (high-efficiency, high-
pressure drop) and toward the middle (mid-efficiency,
mid-pressure drop) in terms of performance, for the media
that 3M can manufacture. As such, these samples should
bracket the performance requirements. 3M has the ability to
customize the media to achieve the required performance for
a given application.
-------
In addition to Freudenberg and 3M, there are many other
electret filter/media manufacturers. The major manufacturers
identified in this project, as well as their contact information,
are listed in Table 4. Table 5 summarizes the performance
Table 4. List of Electret Filter and Media Manufacturers
and cost data collected in the market survey for candidate
electret media. The performance and cost data presented in
Table 5 were either obtained from the manufacturers' product
brochures or provided by the manufacturers directly.
Manufacturer
3M Filtration Products
www.3m.com
Trademark Name of the
Media/Filter
Filtrete™
Types G, B, E, S, and 3M™
AEM
Media Type
Split-fiber, meltblown, etc.
Contact
Filtration Products
3M Center, Building 60-01-S-16
St. Paul, MN 55144
800-648-3550
Ahlstrom Air Media, LLC
www.ahlstrom.com
ELECTROSTAT® HP Series
Triboelectrically charged
Jeffrey Gentry
9319 Cincinnati Columbus Rd.,
Ste. 21
West Chester, OH 45069
513-755-9222, ext.14
Aramid, Ltd.
www.aramid.com
Micron®
NA
Jay Nicholson
24 New Orleans Rd.
Hilton, SC 29928
843-686-2132
DelStar Technologies, Inc.
www.delstarinc.com
DelPore™
Meltblown media
Andrew Platt
601 Industrial Dr.
Middletown, DE 19709
302-378-8888, ext. 4081
Filtrair, Inc.
www.filtrair.com
Filtrair®
Meltblown media
Jay Forcucci
600 Railroad Ave.
York, SC 29745
803-684-3533
Hollingsworth & Vose Co.
www.hovo.com
TECHNOSTAT
Triboelectrically charged
Per Lindblom, Director of Sales
112 Washington St.
East Walpole, MA 02032
501-850-2261
Johns Manville
www.jm.com
Delta-Aire™
HS Series
Meltblown media
Charles R. Granger
171 Sandreed Dr.
Mooresville, NC28117
704-799-1263
Kimberly-Clark Corp.
www.kcfiltration.com
Intrepid®
Continuous Filament Melt-spun
(CFM) media
Kimberly-Clark Filtration Products
1400 Holcomb, Bridge Rd.
Roswell, GA 30076
770-587-8000
Lydall Filtration/Separation
www. lydal I if i Itrat ion .com
LydAir MB
Meltblown Composites
LydAir SC
Synthetic Composites
Scott C. Keeler
North American Sales Manager
Chestnut Hill Rd, P.O. Box 1960
Rochester, NH 03867
603-332-4600, Ext. 155
Toyobo Co., Ltd.
www.tovobo.co.ip
Elitolon®
Types A, AA, U, NA, and FA
Combination of spunbonded and
meltblown fibers
Mitsuhiko Akiyama
AC Department
2-8, Dojima Hama 2 Chome,
Kita-ku, Osaka 530-8230, Japan
+81-6-6348-3372
Freudenberg
www.viledon-filter.com
Viledon®
MV series
MF series
Electrostatic spinning
David J. Matier
Manager of North American
Operations
Freudenberg Nonwovens
Filtration Division
1304 Ramona St.
Ramona, CA 92065
760-788-3833
NA is defined as not available.
-------
Table 4. List of Electret Filter and Media Manufacturers (Continued)
Manufacturer
Camfil Farr
www.camfilfarr.com
Trademark Name of the
Media/Filter
S-Flo
Media Type
Meltblown Synthetic
Contact
Sam Glaviano
Ketchum and Walton Co,
Camfil Farr Representative
1350 W. 5th Ave.
Columbus, OH 43212
Phone: 614-486-5961
3M Filtration Products
www.3m.com
FiltreteTM
H EPA diffuser
Ultra allergen filter
Split-fiber
Filtration Products
3M Center, Building 60-01-S-16
St. Paul, MN 55144
800-648-3550
Due to the limited performance data available, the efficiency
and airflow resistance shown in Table 5 are not based on
the same velocity. Instead, the velocities vary from 10 to 56
fpm, which correspond to 2,000 cfm of air flowing through
media areas ranging from 36 to 200 ft2. This velocity range,
however, covers the operational condition of a typical high-
efficiency, pleated HVAC filter that usually has media areas
ranging from 100 to 180 ft2.
As shown in Table 5, within the velocity range considered,
all the media can provide collection efficiency higher than
84% (for aerosol size <1 um) and airflow resistance less than
0.5 in. of water. Although 84% efficiency for 0.3 um aerosol
is lower than the performance goal, the efficiency for 1 um
could be significantly higher.
In addition to Freudenberg and 3M, Battelle also spoke
with other electret filter/media manufacturers, including
TOYOBO, Kimberly-Clark, Hollingsworth & Vose, and
Lydall. Sample media were requested from the candidate
manufacturers listed in Table 5 and were evaluated in the
candidate media screening tests described in Section 5.0.
4.3.2 Electrically Enhanced Filters
4.3.2.1 Technology Description. An electrically enhanced
filter (EEF) is a technology that can provide bactericidal
activity, relatively high efficiency, and low pressure drop.
The technology has been studied extensively (Bergman et al.,
1983; Jaisinghani et al., 1998). An EEF usually contains an
ionizer for charging the incoming particles, a filter element
for collecting particles, and an electrostatic field across the
filter element for enhancing the collection efficiency.
The operation principle is to ionize the incoming air stream
and particles such that a surface charge is achieved on
the incoming particles upstream of the filter. Charging
these particles will increase both their electrical mobility
as well as the attractive force to oppositely charged
surfaces. Fibrous filter media are located between a
negatively charged electrode upstream and a positively
charged electrode downstream. When power is applied
to the electrodes, an electrical field is generated, and the
fibrous filter media are polarized (i.e., the fibers of the
media form areas of negative and positive charge). In
this manner, it is similar to that of electret media. In the
case of the electrically enhanced filter, the fibers are not
permanently charged like those of electrets, but rather
are charged only in the presence of the electrical field.
Particle collection thus occurs predominantly due to
the electrostatic forces. Because particle collection is
predominantly associated with electrostatic force, larger
fiber diameters of the fibrous filter can be used, allowing
lower airflow resistance. Rather than increase the collection
efficiency of a fibrous filter by reducing the fiber diameter
and thus increasing the pressure drop, the collection
efficiency is enhanced by charging the particles and
polarizing the fibers.
4.3.2.2 Potential Drawbacks. The major drawback
of an EEF system is the potential increase in current
through the filter element when the challenging aerosol
contains conductive particles. For example, 20%
of the ASHRAE test dust (the Arizona Road Dust)
is conductive. The increase in electric current can
automatically reduce voltage and subsequently lead to
a reduction in efficiency. The current increase may also
lead to shorting out of the whole filtration system.
High cost is another disadvantage of the EEF system. The
initial cost of an EEF system is approximately more than 3.5
times the cost of an uncharged high-efficiency filter and more
than 5 times the cost of a high-efficiency electret filter. In
addition, there are extra costs for installation, maintenance,
and operation compared to a conventional filter system.
4.3.2.3 Technology Maturity for Use in the Advanced
Filter. Two commercial EEF systems were identified in
the market survey. The performance data of the two systems
were requested from the manufacturers and are compared
in Table 6. The StrionAir filter (with dimensions of 20" x
24" x 12") was tested by Research Triangle Institute using
the ASHRAE Standard 52.2. The dust-loading test results,
however, were not provided by the manufacturer because of
the degradation due to the electric current increase. An initial
collection efficiency of 95% for 1-um aerosol was measured.
The initial pressure drop at 500 fpm was 0.43 in. H2O.
As shown in Table 6, better performances are claimed for
the Technovation filtration system, which demonstrates
HEPA collection efficiency and 0.5 in. H2O pressure drop
at 600 fpm. However, the current Technovation products
are developed for clean room application rather than HVAC
filtration.
-------
Table 5. Performance Data of Candidate Electret Media from Major Manufacturers
Manufacturer Mode, ^ T, AP (,n. H20> ^ (gj,
3M
AHLSTROM CORP.
DELSTAR TECHNOLOGIES INC.
Filtrair, Inc.
Hollingsworth & Vose Company
Johns Manville
Kimberly-Clark
Lydall Filtration/Separation
TOYOBO
Filtrete™
Type G, G-200
ELECTROSTAT®
HP650/410
ELECTROSTAT®
HP650/410
Del Pore™
DPB002-50PNAT
Del Pore™
DPB002-90PNAT
Filtrair® 95%
TECHNOSTAT
TS 100/1 5
TECHNOSTAT
TS500/15
Delta-Aire™
HS-95
INTREPID 95SP
LydAir
MBCL 1909
LydAir SC
SC8100
Elitolon®
U type, EF-U-98P
_[a]
160
670
50
90
110
115
515
128
-
102
116
105
>99%@40 ft/mi n
(for l|jm)
95% @32 ft/mi n (for
0.1|jm)
99.996% @32 ft/mi n
(for 0.1 Mm)
97% @28 ft/mi n (for
0.3Mm)
99.6%@28ft/min
(for O.SMm)
MERV14 @32 ft/min
>94% @32 ft/min
(for 0.65Mm)
>99.8% @32 ft/min
(for 0.65Mm)
90 to 95%@7 ft/min
(for O.SMm)
90% @28 ft/min (for
O.lMm)
95% @10 ft/min (for
O.SMm)
85% @10 ft/min (for
O.SMm)
84% @20 ft/min (for
O.SMm)
0. 12 @40 ft/min
0.5 @200 ft/min
0.5 @40 ft/min
1 @28 ft/min
0.47 @28 ft/min
0. 08 @10 ft/min
0.03 @40 ft/min
0.31 @40 ft/min
0.5 @51 ft/min
0.5 @48 ft/min
0.1@10 ft/min
0. 5 @56 ft/min
0. 28 @20 ft/min
Both
-
-
Both
Pleat
Both
Pocket
Pocket
Pleat
Pleat
-
-
-
-
-
-
0.07
-
0.05
0.10
-
[aIData are not available from the manufacturer.
Comparing the performance data presented in Tables 5 and 6,
it was found that the collection efficiency (for 1-um aerosol)
and pressure drop of the EEF systems are equivalent to those
reported for the high-efficiency electret filters. The advantage
of an EEF system is that the filter system is claimed to be
bactericidal, due to the combination effects of ionization,
oxidative stress, and current flow across the filter media.
Table 6. Performance Data of Electrically Enhanced Filter Systems
Manufacturer "^ % AP Filter Dimensions Filter Cost
StrionAir
Technovation
Systems, Inc
95% (initial)131 (for l^m)
(@ 1640 cfm or 500
fpm)
> 99.97% (initial) for
O.SMm (@ 2400 cfm or
600 fpm)
0.43 in. H20 (@ 1640 cfm
or 500 fpm)
0.5 in. H20 (@ 2400 cfm or
600 fpm)
20" x 24" x 12"
24"x24"x 12"
EEF system: $800/2000
cfm; plus Disposal filter:
$95/2000 cfm
[b]
[al "OPC and SMPS Efficiency Test Report, StrionAir ElectroFilter," Research Triangle Institute. The report was provided by StrionAir.
[bl Cost data are not available.
-------
Battelle contacted Mr. Rex Coppom, Chief Technology
Officer at StrionAir, to discuss the performance and cost of
StrionAir's EEF filtration system. According to Mr. Coppom,
the MERV 15 performance cannot be sustained when the
EEF is tested following the ASHRAE Standard 52.2 method.
Because 20% of the ASHRAE test dust (the Arizona Road
Dust) is conductive, an increase in electric current through
the filter element occurred during the dust loading, which
automatically reduced voltage and subsequently, efficiency.
Currently, it is not known whether the high conductivity of
ASHRAE dust reasonably represents the conductivity of
ambient aerosol because no literature was found reporting the
conductivity of ambient aerosol (Hanley and Owen, 2003).
StrionAir stated that they are working to improve the EEF
unit to overcome the problem and some progress has been
made. The revised version of design will be submitted for
retesting by LMS and Research Triangle Institute.
The initial cost for the StrionAir EEF system is $800 per
2,000 cfrn of air capacity, which is more than 3.5 times the
cost of uncharged high-efficiency filters and more than 5
times the cost of the high-efficiency electret filters presented
in Table 2. The cost for the disposable filter, which must
be changed every 6 to 12 months on average, is $95 for
2,000 cfm of air capacity.
Generally, an EEF system is much more expensive than an
electret filter system considering the high initial cost and the
additional costs of installation, maintenance, and operation.
Degradation with loading is the other major drawback with
the current version of the technology that prevents it from
providing steady high-collection efficiency. Therefore, the
EEF technology was eliminated from further consideration in
this study as the basis for an advanced filtration system.
4.3.3 Nanofiber Filter
4.3.3.1 Technology Description. Nanofiber filter media
were developed to provide improved filtration efficiency for
a wide range of particles (0.2 to 8 um) without a substantial
increase in pressure drop. "Nanofiber" generally refers
to a fiber with a diameter of less than 1 um. Small fibers
in the nanofiber range can improve filter efficiency in
the interception and inertia! impaction regimes, although
smaller fiber size leads to higher pressure drop. However, a
theoretical analysis conducted by Graham (2002) indicated
that for a fiber size smaller than 0.5 um, the effect of slip
flow at the fiber surface can also lead to better filter efficiency
and lower pressure drop. For air filtration application, small
fiber sizes (0.2 to 0.3 um) are desired.
4.3.3.2 Potential Drawbacks. While nanofiber media
can offer excellent performance in efficiency and airflow
resistance, like any other filtration media, they have
limitations. Based on our discussion with the manufacturers
in the field, nanofiber media are likely more expensive
compared to ordinary fiber media, although the cost data are
not available at this stage. In addition, due to the thinness
of the nanofiber media, their dust-holding capacity (or
service life) could be low, especially when compared to a
conventional deep filter media.
4.3.3.3 Technology Maturity for Use in the Advanced
Filter. The leading manufacturer of nanofiber air filtration
media is the Donaldson Company, Inc. Donaldson makes
polymeric nanofibers using a proprietary electrospinning
process that was developed in the 1970s and has been
enhanced since that time (Barris et al., 2004; Benson et al.,
2004; Chung et al., 2004; Gillingham et al., 2004; Gogins
and Weik, 2004). The Donaldson nanofibers are formed
into a nanoweb, which is very thin—consisting of just a
few nanofiber diameters thick. The thinness of the nanoweb
provides high permeability; however, the nanoweb must be
supported by a substrate material to establish mechanical
properties for use in a filter.
At Donaldson, a variety of substrate materials have been
selected to provide appropriate mechanical properties
to allow pleating, filter fabrication, and durability in
use. In many cases, substrates have been selected that
resemble conventional filter materials to allow the use of
conventional filter media pleating equipment. Donaldson
has used nanoweb technology for a variety of air filtration
applications. The Ultra-Web® filters are used in industrial
air nitrations for dust collection, which demonstrates good
cleanability by pulse-clean. The Spider-Web® filters are used
in gas turbine filtration.
To study the feasibility of using the nanoweb media in
HVAC filtration, a sample medium being developed
at Donaldson for HVAC application was sent to
Battelle for testing. The technical properties of the
sample medium are summarized in Table 7.
Table 7. Technical Properties of Donaldson Nanoweb Media (Grade 1291-20X)[i
Construction
Basis Weight
Corrugation™
Frazier Permeability101
Efficiency
Maximum Operating Temperature
Polyamide nanofibers on a corrugated cellulose/synthetic
blend
120 g/m2
0.013 in.
0.5 in. H20@ 125 fpm
40% on a 0.76 um PSL particle at 20 fpm
200 °F
[al All properties presented in this table were provided by Donaldson Company, Inc.
[bl Filter media thickness.
[cl The measurement of the number of cubic feet of air per minute to pass through a square foot of filter media at a
pressure drop of 0.5 in. of water.
-------
As shown in Table 7, the collection efficiency of the sample
media (as reported by the manufacturer) is significantly lower
than our performance requirement of 99.9% efficiency.
Freudenberg is another manufacturer identified in the market
survey as developing nanofiber filter media for air filtration
application. According to Dr. Manz from Freudenberg,
the nanofiber medium being developed at Freudenberg for
HVAC filtration application has the potential to meet the
performance goals of the advanced filter.
The nanofiber filter technology is still at the early stage
of development for HVAC application. Discussions
with the two major manufacturers in the nano-media
area (Donaldson and Freudenberg) revealed that the
performance goals may be achievable with the technology,
but additional time and effort would be required.
Nevertheless, it was determined that this technology shows
sufficient promise and therefore warranted evaluation.
4.4 Summary of Literature Review
The performance requirements for the advanced filtration
system were established as (a) efficiency higher than 99.9%
for 1-um aerosol and (b) pressure drop less than 0.5 in. H2O
for a filter with dimensions not larger than 24" x 24" x 12" to
handle 2,000 cfm of airflow.
Electret media were selected as the leading candidate
technology to be further evaluated experimentally. The
evaluation of the electret technology would focus on
identifying the minimum collection efficiency using the
method developed by Hanley and Owen (2003).
Samples of the nanofiber media would be experimentally
evaluated to determine the feasibility of improving the
nanofiber media to meet the advanced filtration system
performance requirements. The EEF system was not selected
as a candidate technology to be further evaluated because
of its relatively high cost and the increase in electric current
with dust loading (causing diminished collection efficiency).
-------
-------
5.0
Candidate Media Evaluation Tests
Eleven candidate media (8 electret and 3 nanofiber media)
were obtained from manufacturers noted in Section 4.0
for initial assessment as to whether they would merit
consideration for use in an advanced filter. The sample
media were first tested for airflow resistance and initial
aerosol collection efficiency (or initial penetration fraction).
The sample media were cut into 47-mm diameter circular
swatches and tested in modified commercial 47-mm diameter
filter holders (BGI, Inc.).
Quality factors were calculated for each filter medium
based on the results in airflow resistance and initial aerosol
collection efficiency (details are provided in Section 5.3).
The media with quality factors greater than the performance
goal were selected for further testing to determine
collection efficiency stability. In this test, the test media
were conditioned with a nanometer-sized KC1 aerosol in a
laboratory or exposed to an indoor aerosol, as recommended
by Hanley and Owen (2003), followed by collection
efficiency measurement.
Detailed methods and procedures for airflow resistance
measurement, aerosol collection efficiency measurement,
laboratory conditioning, and ambient aerosol conditioning
are described in Section 5.1. Results of the assessment are
discussed in Section 5.2
5.1 Test Methods and Procedures
5.1.1 Airflow Resistance
The airflow resistance test system is illustrated in Figure 1.
Room air was pulled through the test medium with a vacuum
pump at flow rates corresponding to velocities ranging from
0 to 15 cm/s. The airflow resistance across the medium was
measured with an inclined manometer.
Measurements were first made after increasing the face
velocity from 0 to 15 cm/s and then after decreasing the face
velocity from 15 to 0 cm/s. The average of the two readings
was recorded. A test was also performed at the same range of
face velocities to measure the system pressure drop without
the test media installed in the filter holder. Subtracting the
system pressure drop from those with the test media in line
yielded the net airflow resistance.
Filler
Holder
Manometer
MFM
Pump
Figure 1. Schematic of the Airflow Resistance Test System
-------
5.1.2 Aerosol Collection Efficiency
A schematic of the aerosol efficiency test system is shown
in Figure 2, which consisted of a Collison nebulizer, a Kr-85
neutralizer (Model No. 3012, TSI Inc.), a Climet CI-500 laser
particle counter (Climet Instruments Company), a vacuum
pump, a modified 47-mm filter holder, and mass flow meters.
The Climet CI-500 is designed to detect light scattered by
aerosol particles as they pass through the measuring volume
defined by the width of the instrument's laser beam. To
ensure that only one particle passes through the measuring
volume at a time, the CI-500 has an upper detection limit
of only up to 107 particles/ft3 (350 particles/cm3). This,
however, did not introduce an aerosol counting problem
because the instrument samples the aerosol at a relatively
high airflow rate of 2.83 L/min, and its sampling time was
set to one minute. The size range of the instrument is 0.3 to
10 um, which is broken down into 5 size channels. The data
collected were stored in the unit's internal memory during the
test, after which, they were downloaded into Microsoft Excel,
using the software provided with the instrument.
The nebulizer generated KC1 aerosol in a range of 0.3 to
10 um. The aerosol stream from the generator passed through
a Kr-85 neutralizer. Upon exiting the neutralizer, a portion
of the aerosol stream was pulled through the test filter by a
vacuum pump. The remaining aerosol stream was vented.
During the efficiency test, the Climet CI-500 laser particle
counter was used to size and count the number of particles
upstream and downstream of the test filter media. The ratio
of the downstream counts to the upstream counts was used
to compute the fractional filtration efficiency for each of the
particle size channels.
The tests were performed with a single layer of the test
medium at velocities of 6.8, 10.2, and 13.6 cm/s. These
three velocities correspond to 2,000 cfm of air flowing
through media areas of 150, 100, and 75 ft2, respectively,
which cover the typical media areas in a 12-inch (30 cm)
deep pleated commercial high-efficiency HVAC filter. After
initiating the aerosol challenge at 6.8 cm/s, the particle
counts in each size channel were recorded upstream for
approximately 4 minutes. The particle counts were then
recorded downstream for the next 4 minutes and then
upstream again for another 4 minutes. This procedure was
then repeated at 10.2 and 13.6 cm/s. Note that the challenge
aerosol concentration was relatively stable during the test
period, with typically less than 15% change in the upstream
particle counts (for every channel) during a 15-minute testing
time. To eliminate measurement error due to any unstable
challenge concentration, at each test flow rate, upstream
aerosol concentration was measured before and after every
downstream concentration measurement. The average of
the two upstream concentrations was then used with the
downstream concentration to calculate the filter penetration.
To eliminate test system bias, background aerosol
concentrations upstream and downstream of the test filter
were measured at the beginning of each test with the aerosol
generator turned off and with clean air flowing through the
test filter at the testing flow rate. The penetration (p) was then
calculated based on the ratio of the downstream to upstream
particle concentrations corrected on a channel-by-channel
basis as shown in Equation 1:
p =•
(cD-cDb)
(1)
where: CD = Downstream particle count, particles/cm3,
CDb = Downstream background count, particles/cm3,
Cv = Average upstream count, particles/cm3, and
Cub = Average upstream background count,
particles/cm3.
The collection efficiency TJ was then computed, as shown in
Equation 2:
,/(%)= 100 x(l-p) (2)
As illustrated in Figure 2, aerosol concentration
was measured with the Climet CI-500, using
identical sampling probes positioned upstream and
downstream of the filter holder. The sampling ports
were located approximately 5 cm from the filter holder
on the 1.2-cm diameter inlet and outlet tubes.
Note that some percentage of particle loss (especially for
larger particles) through a sampling probe is inevitable in
sampling probe design. However, as illustrated in Equation 3,
the measured filter efficiency can be corrected if the particle
deposition efficiencies through the sampling probes are
known.
r Measured Downstream Cone, -i
—
Filter Collection Efficiency (%) = -
r Measured Upstream Cone, -i
(3)
(1 — T! dsp _ upstream)
where rjdep_domstream and rjd^upamamWQ the particle deposition
efficiencies through the downstream and the upstream
sampling probes, respectively.
Using a pair of identical probes in each measurement
eliminated the need to know the particle deposition
efficiency; since the upstream and downstream deposition
efficiencies are identical, they can be cancelled out from
Equation 3.
-------
0
o
LO
u
tu
E
u
(1)
(J
•-F
_c
o
CO
C
-Q
U LL.
-------
5.1.3 Laboratory Conditioning and Efficiency Evaluation
The laboratory conditioning was conducted using the
conditioning procedure described by Hanley and Owen
(2003). Hanley and Owen's study was supported by
ASHRAE to establish a laboratory conditioning method to
identify the minimum efficiency of electret filters in actual
use. At the time of conducting this study, the conditioning
method developed by Hanley and Owen was in public review
and is to be included as an addendum to ASHRAE 52.2 for
electret media evaluation.
The conditioning aerosol was generated using a Collison
nebulizer. The nebulizer was operated with an aqueous
solution of 0.03% KC1. The number concentration of the
conditioning aerosol was monitored with a PORTACOUNT®
condensation nucleus counter (TSI Inc., Shoreview, MN).
Particles entering the PORTACOUNT® pass through
a saturator tube where they are combined with alcohol
vapor. They then pass into a condenser tube where alcohol
condenses on them, causing each particle to grow into a
larger droplet. The droplets then pass through a focused
laser beam, producing flashes of light that are sensed by a
photodetector. The particle concentration is determined by
counting the light flashes. The PORTACOUNT® performs
sampling at a flow rate of 0.7 1pm. During the conditioning,
the aerosol number concentration was kept below IxlO6
particle/cm3 to prevent excessive coagulation of the
conditioning aerosol.
At the beginning of all tests, the size distribution of the
laboratory conditioning aerosol was characterized with
a Scanning Mobility Particle Sizer (SMPS, TSI Inc.,
Shoreview, MN). This instrument consists of an electrostatic
classifier, which is used to separate particles by size, and a
condensation particle counter, which counts the particles. The
detection range of the instrument is 20 to 107 particles/cm3.
In the test configuration, this instrument had a sample rate
of 0.3 L/min and a sample time of 2 minutes. The design of
the instrument is such that the particles are counted one size
channel at a time; thus, each size channel is sampled for only
a fraction of the 2-minute sampling time. The effective range
of particle diameters measured by the instrument was 0.015
to 0.66 um. The WPS was controlled by a computer, and
all data were collected using the TSFs "Aerosol Instrument
Manager" software. The data collected were then transferred
to Microsoft Excel, using the "cut and paste" function, for
further analysis. The size distribution was not monitored
during the conditioning.
To identify the minimum efficiency of the electret sample,
conditioning was performed in incremental steps with
efficiency measured after each increment. A manifold
consisting of six 47-mm filter holders was fabricated, and six
sample swatches were conditioned each time. The collection
efficiencies of the conditioned test media were then measured
one at a time after each incremental conditioning step with
0.3 to 10 um KC1 aerosol at 10 cm/s. The 10 cm/s face
velocity corresponded to 2,000 cfm air flowing through an
overall media area of 100 ft2. The efficiency measurement
procedure was identical to that described in Section 5.1.2
except that the measurement was made at only one face
velocity of 10 cm/s instead of three face velocities.
5.1.4 Ambient Conditioning and Efficiency Evaluation
As illustrated in Figure 3, the ambient conditioning test was
conducted by exposing test medium to indoor aerosol at 10
cm/s in incremental steps. The collection efficiency of the
conditioned media was measured after each incremental
conditioning step with 0.3 to 10 um KC1 aerosol at 10 cm/s.
At the beginning of the ambient exposure test, the size
distribution of the indoor ambient aerosol was characterized
with a Wide-Range Particle Spectrometer (WPS, Model
1000XP, MSP Corp. St. Paul, MN). The WPS is based
on the same principle as that of the SMPS, with aerosol
sizing and counting by laser light scattering, differential
mobility analysis (DMA), and condensation particle
counting (CPC). The detection range of the instrument
is 20 to 107 particles/cm3. In the test configuration, this
instrument had a sample rate of 0.3 L/min. and a sample
time of 1.6 min. The effective range of particle diameters
measured by the instrument was 0.015 to 0.5 um. The
SMPS was controlled by a computer, and all data were
collected using the manufacturer-provided software.
Ambient
Indoor Air
47-mm
Filter Holder
Wide-Range
Particles
Spectrometer
Mass Flow Valve
Meter
Pump
Figure 3. Schematic of the Ambient Conditioning
Test Setup
-------
Table 8. Test Matrix
Number of Tests
Media Type Media ID Airflow Resistance Initial Penetration Laboratory Ambient
(0 ~ 16 cm/s) (@ 6.8, 10, 13.5 cm/s) Conditioning Conditioning
Electret Media
Nanofiber Media
A
B
C
D
E
F
G
H
1
J
K
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
_[a]
-
-
-
2
5
-
-
-
-
-
-
-
-
-
-
2
-
-
-
-
[aITest was not conducted at the condition.
The instalments and equipment used in the test were
calibrated according to the manufacturer's recommendations
or on an annual basis. All digital flow meters were
calibrated by the Battelle Instrument Lab following standard
procedures. The particle-counting instruments, including
Climet CI500, SMPS, PORTACOUNT®, and WPS, were
calibrated according to the recommended instrument
calibration frequencies and procedures provided in the
respective manufacturer's manuals. All tests were conducted
at ambient temperature and relative humidity. The particle-
counting instruments were used within the detection limit of
particle concentrations recommended by the manufacturers to
ensure the calibrated precisions were achieved.
5.2 Test Results and Discussions
Tests were conducted to evaluate the airflow resistance and
initial particle penetration of all candidate media. The top
three media (all electret) were then further evaluated for
efficiency stability after conditioning in laboratory or indoor
ambient air. The test matrix is presented in Table 8.
The test results are presented in Sections 5.2.1 to 5.2.5
for airflow resistance, initial penetration fraction, quality
factor, laboratory conditioning, and ambient conditioning,
respectively. The sample media are designated as Samples A
through K. To comply with the non-disclosure requirements
from some media manufacturers, no further identification
regarding media manufacturer, media type, or media
properties is provided.
5.2.1 Airflow Resistance
The initial airflow resistances of the candidate media were
measured at velocities ranging from 0 to 15 cm/s. Figures 4
and 5 present the results of the electret media and nanofiber
media, respectively. Duplicate tests were conducted for each
candidate material. The pressure drops plotted in Figures 4
and 5 are the average of the two tests; the error bars provide
the range of the two measurements.
As expected, the pressure drop increased linearly with
velocity in the range of velocities studied. As the velocity
increased from 6.8 to 13.5 cm/s (equivalent to reducing the
design media area of a 2,000-cfm filter from 150 to 75 ft2),
the airflow resistance increased by approximately a factor of
two for all sample media tested.
-------
25
20
O
f 15
10
£
PM
^
• Sample A
0 Sample B
• Sample C
0 Sample D
° Sample E
• Sample F
o Sample G
• Sample H
|
x
X-
«g
x
^
^
,,
^
-^
s-
x^
x^
^^
^
x
^^
fi
/
X
^^
=-
X
'^
/
^x
^
^
^
^
=8
-°
-•
x
^
^
^
—
x
^
^
^
"
- —
x
^
'
^
^~-
~~
x
X
^
^
— -
^
,,
^
--
- — '
X
X
^-
—
^
"*'
'
_J(!
'rv
a
_»-
x
x
**
-
•
x
^
-*-
— -
x
x
5S*
^-
— '
^
—
x
x
=^
--
—
x
x
^
.
^
• — "
X
x
S5*
_,
^-
if
»-
*
.0-
«'
tT
X
-^
^
-^
-^'
x'
x
#=
^
x
-^
'-^
^
^-
x
x
^;
--
-"
^
x
x
i==
-^
--
-x
^x
-^
-^"
^x
•£
If-
a~
a-
r
x
x
^
-"
—
__^,
X
x
-^
"
'
_-
x
x
X
.
— '
"
x
x"
^
--
^^
"
x
x
^
^
--
x^
•^
-d
-^^
^-E
1
i
»
i
^
i
1
0
4
12
6 8 10
Velocity (cm/s)
Figure 4. The Initial Airflow Resistance of Candidate Electret Media
14
16
18
ou -
O'40
S
a-
Pressure Dr
.> u
2 C
^
x
/
=s==is
0
0
0
X
^^
Sample 1
Sample }
Sample K
x
-^
^^
/
a=
X
=^
x
•
x
X
^
^
s:
x
—
xB
-J
X
-—
x
^
X
X
,
—
x
-^
S3-
Q
-*-
x
____
x
-_-
— "
X
- —
_,
x
_
.
X
_—
- —
-lX
1i
4-
*-
x
:
X
x
_-^-
. -
x
-—
—
X
~~~
,,
tf
f-
X
—
X
"
x
x
X
-—
"^
1
1
1
0 2 4 6 8 10 12
Velocity (cm/s)
Figure 5. The Initial Airflow Resistance of Candidate Nanofiber Media
14
16
18
5.2.2 Initial Penetration Fraction
The initial penetration fractions of the electret candidate
media at 10 cm/s are compared in Figure 6. As expected,
the penetration fraction decreased as the particle diameter
increased. For different electret, the measured penetration
fraction varied within a large range. For example, at 0.75 um,
the measured penetration fraction ranged from 0.0001 for
Sample G to 0.26 for Sample E. Similarly, as shown in
Figure 7, the penetration fraction of the nanofiber media
also increased as the particle size decreased. At 0.75 um, the
measured penetration fraction of the nanofiber media ranged
from 0.0005 for Sample K to 0.13 for Sample I.
-------
0.1
2 o.oi
o.ooi
0.0001
4
I
|
I
I
^\
^\
\V
\
\
\»
^
n\
•
\
\
&
-X
\.
^-
\ \
\
\
\\
\
^
\>
•
"vi
IS
c
-
lat
'ek
ha
oia
—-
eri
5Ci1
Her
lyz
^=^
il:
y
ige
er:
=
Ele
10
K
Cl
^
— -
ctr
cm
Cl
me
^=
~~
-~^_
3tl\
Is
Ae
tc
__
-__
dec
ros
lia,
ol(
00
-•
-•
— n
Sil
0.3
La
~~~
~_^^
igle Layer
to lOmicrc
5er Particle
-—-
—
-~^,
-~^_
~~~~-
-~-
"~~~-
-^
n)
Counter
\
^^^
~—
~-
~~~~-
-~
^
--
^
\
-~
\
—
~~*~ Sample A
-O- Sample B
-•- Sample C
• Sample D
— •— Sample E
• Sample F
— •— Sample G
• Sample H
-_
—
—
—
^
—
^
•
=•-
r*
0
1
2345
Particle Diameter (micron)
Figure 6. Initial Penetration Fraction of Electret Candidate Media
i
o.i
0.01
0.001
1
<
<
1
\
r\
\
1
\
\
\
\
\
\
\
•
\
\
\
•
\
\
\
\
•
•
^
__
-
- —
-•
— •
~^;
^
~~~~~
(
I
__~
vlaterial: Nano Media, Single Layer
/elocity: lOcm/s
Challenge: KC1 Aerosol (0.3 to 10 micron)
Analyzer: Climet CI-500 Laser Particle Counter
- — -
~-
-
—-
_^_
~~—
- — -
-— -
.
_____
~~
"~~^~-
^>
H>
^>
—
-~
-s
-s
-s
—
~~~
am
am
am
— _
-^
pie
pie
pie
-^
I
J
K
=•-
-•-
0.0001
012345
Particle Diameter (micron)
Figure 7. Initial Penetration Fraction of Nanofiber Candidate Media
To investigate the effect of velocity on collection efficiency,
the initial efficiency of the candidate media was also
measured at two other velocities (6.8 and 13.5 cm/s). The
results are presented in Figure 8. As shown in Figure 8, the
effect of velocity on initial penetration of 1-um particles
is not significant. For all candidate media tested, when the
velocity increased from 6.8 to 13.5 cm/s, the change in
efficiency of 1-um particles was less than 7%. Therefore,
increasing media area in a filter design can effectively reduce
pressure drop; however, its benefit to collection efficiency is
limited, within the range of velocities tested.
-------
0.00010
0.00001
i.ooooo
• o.ioooo
: 0.01000
; o.ooioo
a o.oooio
0.00001
j
-4-
)
(
!
|
]
T-
c
A
!edia: Sample A, Electret
hallenge: KC1 aerosol (0.3 to 10 micron)
Jialyzer: Climet CI-500
6.8 cm/s
10 cm/s
234567
Particle Diameter (micron)
Figure 8a. Sample A
• 6.8 cm/s
• 10 cm/s
1
• 13.5 cm/s -
Media: sample c, Jilectret
Challenge: KC1 aerosol (0.3 to 10 micron)
Analyzer: Climet CI-500
f
;
l.UUUUU
a 0.10000
i
a 0.01000
o
§ 0.00100
.1
c
~ " 0.00010
n 0.00001
8
1.00000 -i
0.10000
o
B 0.01000
o
| 0.00100
s
" 0.00010
Media: Sample B, Electret
Challenge: KC1 aerosol (0.3 to 10 micron)
Analyzer: Climet CI-500
4
c
*
0
.
S
1
T
A
T
2345
Particle Diameter (micron)
Figure 8b. Sample B
6.8 cm/s
1
Media: Sample D, E
Challenge: KC1 aert
Analyzer: Climet C
le
5
f
I
ctret
1(0. 3 to 10 micron)
00
• 6 8 c
- • 10 cr
- o 13.5
6 7
n/s ; =
cm/s --
T
8
II
— -
2345
Particle Diameter (micron)
Figure 8c. Sample C
2345
Particle Diameter (micron)
Figure 8d. Sample D
1.00000
. 0.10000
- 0.01000
0.00001
(
1
- • 6.8 cm/s -
• 10 cm/s
}
- J-
vledia: Sample E, '.
hallenge: KC1 aer
Analyzer: Climet C
^le
I-
|
ctret
ol(0.3to 10 micron)
00
I
a 0.10000
o
S
B 0.01000
o
b
| 0.00100
.s
a
" 0.00010
ii nnnm
1
- • 6.8 cm/s --
= • 10 cm/s EE
1
--•
VI
a
\n
T
dia: Sample F,
alien ge: KC1 ae
alyzer: Climet C
31
o
I-
i
sctret
ol (0.3 to 10 micron)
500
1
2345
Particle Diameter (micron)
Figure 8e. Sample E
2345
Particle Diameter (micron)
Figure 8f. Sample F
Figure 8. Initial Collection Efficiency of Candidate Media at Varying Velocities
-------
l.llllllllll
0.10000
0.01000
( 0.00100
3
'S
" 0.00010
!
f
0
a 0.10000
B 0.01000
| 0.00100
*s
" 0.00010
•
'
0
|
8
i
i-
A
edia: Sample G, Electret
lallenge: KC1 aerosol (0.3 to 10 micron)
nalyzer: Climet CI-500
2345
Particle Diameter (micron)
Figure 8g. Sample G
n
---1
6.8 cm/s
.0 cm/s
13.5 cm/
J
s
c
A
— • 6.8 cm/s -
• 10 cm/s
6
^edia: Sample I, Nano-fiber
hallenge: KC1 aerosol (0.3 to 10 micro
jia yzer: Climet CI-500
012345
Particle Diameter (micron)
Figure 8i. Sample I
a 0.10000
1
B 0.01000
.s
1
C
;
•
i)
7
|
6 7
i
*
«
•_
tt
c
«
"_
0.100
0.010
111
c
8
>
Penetrati
0
!t
c
8
= Ana
• 6.8 cm/s
o.oooio —
- • 10 cm/s
0.000
1.000
0.100
0.010
0.001
0.000
0.000
a:
lei
yz
S
er
an
C
0 13.5 cm/s
0 1
Q
)0 i
M I
0 1
pie K, Nano-fiber
KC1 aerosol (0.3 to 10 mi
limet CI-500
Cha
Ana
:::::::::!
: |=
^^^t
=F
ia: Sample H, Electret
lenge: KC1 aerosol (0.3 to 10 micron
lyzer: Climet CI-500
2345678
Particle Diameter (micron)
Figure 8h. Sample H
• 6.8 cm/s Media
• 10 cm/s Chalh
n , -, - , Analy
: Sample J, Nano-fiber
nge: KC1 aerosol (0.3 to 10 micron)
zer: Climet CI-500
;
2345678
Particle Diameter (micron)
Figure 8j. Sample J
s o.ooioo ;
0.00010
01234567
Particle Diameter (micron)
Figure 8k. Sample K
Figure 8. Initial Collection Efficiency of Candidate Media at Varying Velocities (Continued)
-------
5.2.3 Quality Factor
To compare the overall performance of candidate media, a
parameter designated as "quality factor" was introduced. As
shown in Equation 4, the quality factor (QF) is defined as:
QF =
APx8
(4)
where: p is the penetration fraction of 1-um particles at
10 cm/s,
AP is the pressure drop (mmH2O) at 10 cm/s, and
5 is the filter media thickness (mm).
Note: 25.4 mm = 1 inch
The filter media thickness (5 ) was measured by Battelle. For
each candidate media, three filter samples were measured.
The average thickness was used in Equation 3 to calculate
the QF. For a medium that meets the current performance
goal, the QF is 0.54. Table 9 summarizes the QF, the
initial penetration fraction, the initial airflow resistance,
and the media thickness of all candidate media tested. For
comparison, the target QF based on the performance goal is
also presented in Table 9.
As shown in Table 9, all three nanofiber media (Samples I,
J, and K) had QFs much lower than the target quality factor.
With the QFs all lower than the baseline QF of 0.54, this
indicates that any advantage the media has with regard to
higher efficiency or lower pressure drop than the requirement
is more than offset with a corresponding lower efficiency or
higher pressure drop. Therefore, none of the nanofiber media
were further evaluated for collection efficiency stability.
Among the eight candidate electret tested, only Samples
A, F, and G have QFs higher than 0.54. This means that
the media offer a combined increased efficiency or reduced
pressure drop compared to the baseline requirement that is
beneficial to overall filter performance. These three sample
media were selected for further conditioning tests to evaluate
their potential degradation with aerosol loading. Candidate
media F and G are from the same manufacturer, and when
the samples were provided, the manufacturer indicated that
they could further engineer the media. Samples F and G were
intended to bracket the target filtration efficiency and airflow
resistance, with the intent that a medium could then be
engineered to more closely meet target specifications.
5.2.4 Laboratory Conditioning
Based on the quality factor, the initial efficiency, and the
airflow resistance, sample media A, F, and G (all electret
technologies) were selected as the most promising media for
further evaluation with laboratory aerosol conditioning.
At the beginning of the tests, the size distribution of the
conditioning aerosol was characterized with a TSI Model
3080 Scanning Mobility Particle Sizer (SMPS, with TSI
Model 3081 DMA and Model 3025 CPC). The SMPS
operated at an impactor inlet diameter of 0.0457 cm, a
sample flow rate of 0.3 1pm, a sheath flow rate of 3 1pm, and
a size range from 15.1 to 661 nm. The results are presented in
Figure 9. The challenge conditioning KC1 aerosol is in solid-
phase, with a particle density of 1.98 g/cm3. The number
mean diameter was approximately 34 nm, with a geometric
standard deviation of approximately 1.55. The mass mean
diameter was 67 nm. Note: Count mean diameter (CMD)
which is shown on the graphs is synonymous with number
mean diameter.
During the conditioning, the aerosol number concentration
was measured periodically with a TSI Model 8020
PORTACOUNT®. The average number concentration was
approximately 400,000 particles/cm3. The aerosol collection
efficiencies of the sample media were measured after each
conditioning stage. Duplicate tests were conducted for each
candidate media. The average penetration fraction of the
duplicate tests is presented in Figures 10, 11, and 12.
Table 9. Quality Factor Comparison of the Candidate Sample Media (At 10 cm/s for 1-um particle)
Media Type Sample I.D. Quality Factor AP (mmH20) Thickness (mm)
Electret Media
Nano-Media
Sample A
Sample B
Sample C
Sample D
Sample E
Sample F
Sample G
Sample H
Sample 1
Sample J
Sample K
Performance Goal
0.59
0.27
0.31
0.50
0.36
1.20
0.86
0.44
0.22
0.26
0.26
0.54
8.9
6.8
9.0
5.0
6.1
2.8
13.5
11.0
10.1
12.0
34.8
12.7
0.044
0.0001
0.063
0.172
0.211
0.095
0.0001
0.055
0.11
0.059
0.0003
0.001
0.6
5.1
1
0.7
0.7
0.7
0.8
0.6
1
0.9
0.9
1
-------
10000
7500
5000
2500
0
Aeros<
}1: Nan
3-sizs
;KC
:i>
Ler
OS
3l
fo
r
Electret Conditioning
Analyzing Equipment: SMPS
- CMD = 34 nm
-I
v!MD =
3EO. S
67 n
td. D
m
ev. =
= i
55
i
6
&
0?
o
°0°
8>
o
OQ
J^
g?^
»
>
0
o
*
0
(8>
o
®
(
o
!
>
o
i
s>
1
i
1
™
fei
%
1
^
8
^taano
o Sample 1
• Sample 2
o Sample 3
1 10 100
Diameter (nm)
Figure 9. Size Distribution of the Conditioning Aerosol (Measured by SMPS)
1000
1.0000
0.1000
§ 0.0100
0.0010
0.0001
~ Sample A, initial
- Sample A, after 2 hours
~ Sample A, after 5 hours
- Sample A, after 8 hours
- Sample A, after 13 hours
345
Particle Diameter (micron)
Figure 10. Incremental Laboratory Conditioning of Sample A Electret Media
-------
1.000
0.100
0.010
0.001
\
Sample F, initial
~ Sample F, after 2 hours
~ Sample F, after 5 hours
Sample F, after 8 hours
Sample F, after 13 hours
345
Particle Diameter (micron)
Figure 11. Incremental Laboratory Conditioning of Sample F Electret Media
0.1000
0.0100
0.0010
~ Sample G, initial
~ Sample G, after 2 hours
- Sample G, after 5 hours
- Sample G, after 8 hours
- Sample G, after 13 hours
- Sample G, after 24 hours
0.0001
345
Particle Diameter (micron)
Figure 12. Incremental Laboratory Conditioning of Sample G Electret Media
As shown in Figure 10, Sample A degraded significantly
during the first 2 hours of conditioning. At 0.75 um, the
penetration fraction increased from the initial 0.06 to 0.23
after 2 hours of conditioning. The penetration fraction then
remained at 0.19 after 5 and 8 hours of conditioning, and
reduced to 0.1 after 13 hours of conditioning, indicating
the maximum penetration fraction of Sample A was
approximately 0.23.
Sample F media degraded continuously during the 13 hours
of conditioning. As shown in Figure 11, the penetration
fraction at 0.75 um increased from the initial 0.12 to
0.31, 0.51, 0.57, and 0.67 after 2, 5, 8, and 13 hours of
conditioning, respectively. Compared to Samples A and F,
Sample G demonstrated excellent stability in penetration
fraction. The penetration fraction at 0.75 um increased only
0.0013 from the initial 0.0001 to 0.0014 after 24 hours of
conditioning.
-------
To verify the penetration stability measured for Sample G,
the laboratory conditioning test of Sample G was repeated.
To identify the maximum penetration of Sample G media, the
overall conditioning time was extended to beyond 24 hours.
Triplicate tests were conducted. The average penetration
fraction of the triplicate tests is presented in Figure 13.
Similar to that demonstrated in Figure 12, in the
repeated test (see Figure 13), the reduction in penetration
fraction was not significant after 35 hours of laboratory
conditioning. The penetration fraction (the average
of three sample swatches) as a function of time is
presented in Figure 14, where the error bars illustrate the
deviation of the individual sample from the average.
As shown in Figure 14, the penetration fraction reached
its maximum penetrations of 0.013 and 0.0015 for
0.4 and 0.75 um particles, respectively, after 15 hours
of laboratory conditioning. Note that the minimum
efficiency for 1-um particles (interpolated using
the efficiencies at 0.75 um and 1.75 um) just met
the performance goal of not lower than 99.9%.
The airflow resistances were measured after the final
conditioning of the three candidate electret media. The results
are presented in Figure 15. The airflow resistance of Sample
G media more than doubled after 35 hours of laboratory
conditioning, implying that the voids between electret fibers
were reduced and the mechanical-collection mechanism
became important.
o.iooo
o.oioo
o.ooio
o.oooi
o.oooo
1
<
1
\\\
\\
\\\
1
\
\
\
\
\
\
\
\\\
u
y.
n
L
•
\
\
^
\
\
_X
\
\ '
\\
V
\
^ \
\\ \
\
\
\\\\
\
\
1
A\
\
lb
—
— -
—
-o- Sample G, initial
-•- Sample G, after 2 hours
-•- Sample G, after 7 hours
~*~ Sample G, after 15 hours
~*~ Sample G, after 25 hours
-o- Sample G, after 30 hours
-•- Sample G, after 35 hours
345
Particle Diameter (micron)
Figure 13. Incremental Laboratory Conditioning of Sample G Electret Media (Repeated Test)
-------
0.1
0.01
=
netration Fract
I ?
J C
J C
i h-
3 u.uuui
PLH
0.00001 <
s
/
s
I
/
/
i
' I/
v
/
/
/
/
/
-r"
.•I
^ j
i
^~^— <
— •• — 0.4 micron particles
— o — 0.75 micron particles
. i
^^_
i ^
"** -^
^i
Xj
0
35
40
5 10 15 20 25 30
Laboratory Conditioning Time (hours)
Figure 14. Penetration Fraction as a Function of Conditioning Time (for Sample G Electret Media)
60 i
40
0
O Sample A, initial
A Sample A, after 13 hrs conditioning
O Sample F, initial
A Sample F, after 13 hrs conditioning
O Sample G, initial
A Sample G, after 24 hrs conditioning
* Sample G, after 35 hrs conditioning
12
14
2 4 6 8 10
Velocity (cm/s)
Figure 15. The Airflow Resistance of Candidate Electret Before and After Conditioning
16
18
-------
5.2.5 Ambient Conditioning
The laboratory conditioning tests presented in Section
5.2.4 identified Sample G as the most promising electret
with the best stability in collection efficiency. As discussed
in Section 4.3, the collection efficiency stability is the
most important parameter for assessing an electret filter
medium, which determines its minimum collection
efficiency. Electret media with higher efficiency stability
would have a higher minimum collection efficiency
when compared to an electret medium with the same
initial efficiency but lower efficiency stability.
Sample G was tested further for efficiency stability with
ambient conditioning. At the beginning of the ambient
conditioning test, the size distribution of the indoor aerosol
was characterized with a Wide-Range Particle Spectrometer
(WPS™, Model 1000XP, MSP Corp., St. Paul, MM). The
WPS was selected to characterize the size distribution of
the conditioning indoor aerosol because indoor aerosol
contains a significant amount of fine particles at nano-
size range and the WPS (like the SMPS) is able to detect
particles with diameters down to 0.015 um. Compared to
SMPS, the WPS operates based on the same aerosol sizing
and counting principles, and is able to detect the similar
particle size range with the same detection limit in total
particle concentration as that of the SMPS. Therefore, WPS
is expected to provide equivalent results in size distribution
and particle concentration to those of the SMPS. The results
are presented in Figure 16. The number mean diameter was
approximately 67.2 nm, with a geometric standard deviation
of approximately 1.9. The mass mean diameter was 206 nm.
The indoor aerosol distribution stability was not monitored
during the long-term conditioning.
The aerosol number concentration was measured periodically
with a TSI Model 8020 PORTACOUNT® during the ambient
conditioning test. The average number concentration was
approximately 13,300 particles/cm3.
300
250
200
3 150
100
50
0
.
1
1
o
o
o
0
3
1
O
o
1
3
0
c
f
1
1
1
0
o
c
3
C
3
C
0
o
II
II
Aerosol: Ambient Aerosol for Electret Media
Ambient Conditioning
Analyzing Equipment: WPS (MSP, Model 1000-XP)
- CMD = 67.2 nm
I
3
0
o
•
0
§
- MMD = 206 nm
- GEO. Std. Dev. =
1
°8
•o-
o
-o
1.9
^o
• Sample 1
• Sample 2
o Sample 3
-o-o-o
-o-o-
•o-o
-o-
-o
-o-
o
or
1
10
1000
100
Diameter (nm)
Figure 16. Size Distribution of the Indoor Ambient Conditioning Aerosol (Measured by WPS)
10000
-------
Duplicate samples were conditioned and the aerosol
penetration of each was measured. The average
penetration fraction measured is presented in Figure 17.
As shown in Figure 17, after 16 and 34 days of ambient
conditioning, the aerosol penetration fraction increased
by factors of ~2 and ~4, respectively, for 0.75-um
particles. The penetration fraction increased 30% for
1.75-um particles after 34 days of conditioning. The
penetration at 1 um met the performance goal of less
than 0.001 during all tests. The penetration at 1 um was
interpolated using data at 0.75 and 1.75 um, assuming
the penetration fraction curve is approximately linear at
particle diameters from 0.75 to 1.75 um. This assumption
of a pseudo-linear curve is reasonable, considering the
relatively small interpolation interval in diameter.
5.3 Summary of Experimental Study
Based on the screening tests (airflow resistance and initial
collection efficiency) conducted with sample media from
the candidate manufacturers, none of the nanofiber media
met the performance goals of the supposed advanced
filtration system.
One electret media, Sample G, demonstrated high potential
to meet the performance goals. The initial and minimum
collection efficiencies met the performance goal of > 99.9%
(penetration fraction of < 0.001) for 1-um diameter particles.
The initial pressure drop of 0.53 in. H2O was slightly higher
(6% higher) than the requirement of the advanced filtration
system. By enhancing the filter design area slightly (over
100 ft2), the performance goal in airflow resistance could
most likely be met.
The manufacturer of Sample G indicated they are developing
a high-efficiency electret filter with a target performance
of MERV 16. The new electret filter uses an improved
electret media that has even better efficiency stability (over
time) than Sample G. According to the manufacturer, the
new electret filter is being tested and will be commercially
available soon.
5.4 Advanced Filter Development
The original objective of the project discussed in this report
was to develop an advanced filtration system and then assess
its performance using the ASHRAE 52.2 test method. In
review of the results presented in Sections 5.1 through 5.3
above, it was considered possible to develop a filter with
better performance than that of filters currently on the
market. It was determined, however, that the incremental
gain in collection efficiency, along with the incremental
reduction in airflow resistance, were not sufficient to merit
continuing with the development of the advanced filter under
this project. In addition, the decision to not proceed with
developing the advanced filter in this project was also due to
the fact that the manufacturer of the leading media assessed
(Sample G) was already making further improvements to
that material and a filter made of the improved media was
expected to be on the market soon.
o.oioo
o
\
\
\
1
I
\ \
\ \
\ \
\\
Zu
I \\
Ux
\ \
\
\
\
\
n
\
, \
\ •—
-fe ,
_____^
~~*~ Sample G, Indoor Ambient Conditioning, initial
~~*~~ Sample G, Indoor Ambient Conditioning, after
~°~ Sample G, Indoor Ambient Conditioning, after '
— n
6 days
4 days
o
0.0001
012345
Particle Diameter (micron)
Figure 17. Indoor Ambient Conditioning of Sample G Electret Media
-------
6.0
Conclusions and Recommendations
A literature search and market survey were conducted to
identify candidate advanced filtration technologies that could
be used as the starting point for further developing a filtration
system that has a lower pressure drop than conventional
high-efficiency paniculate filters, with higher or equivalent
collection efficiency and comparable or lower cost. As
a result of the literature review and market survey, two
technologies (electret and nanofiber media) were identified as
potential candidate technologies to be used for the advanced
filtration system. Sample electret and nanofiber media were
obtained from manufacturers and tested to explore the
feasibility of developing the advanced filtration system.
To evaluate the candidate technologies, performance goals
were established for the advanced filtration system based on
the following criteria: (a) has better performance than the
high-efficiency filters (MERV 14, 15, and 16) available in the
market and (b) does not exceed the pressure drop limit that
common HVAC systems can accommodate. The performance
goals were thus established as follows: 99.9% efficiency for
1-um particle and a pressure drop of less than 0.5 in. H2O.
Three candidate nanofiber media with different levels of
target efficiencies were tested, and the results showed their
performance to be significantly lower than the performance
goals. Therefore, the nanofiber media were excluded from
further consideration for use in an advanced filtration system.
From the tests conducted on the eight candidate electret
media (from seven manufacturers), only one (Sample G)
demonstrated both collection efficiency and airflow resistance
close to the performance goals. Both the initial and the
minimum collection efficiencies of Sample G (the latter
was determined after laboratory conditioning) can meet the
efficiency goal of 99.9% for 1-um diameter particles. The
initial airflow resistance, however, was approximately 6%
higher than the performance goal. The slightly higher airflow
resistance can most likely be reduced by enhancing the filter
media design area to over 100 ft2, which is attainable since a
typical high-efficiency HVAC filter (pleat) usually has media
areas ranging from 100 to 180 ft2.
In conclusion, the test results with Sample G media showed
the potential to develop an advanced electret filter that can
meet the performance goals. However, improved electret
media, with more stable collection efficiency (over time)
than that of the Sample G media are already being developed
by the manufacturer. These new media are expected to be
commercially available soon. For these reasons, no further
development of an advanced filtration system was performed
in this project.
-------
-------
7.0
References
3M Company. "Breathe Easier Filtration," 3M Brochure, www.mpsupplies.com/3miiltreterilters.pdf.
AAF International, "Anthrax FAQ's," www.aafintl.com. 2005.
AIRGUARD, "Protecting Buildings from Bio-Terrorist Attacks," www.airguard.com. 2004.
Arnold, B. and D. Myers. "Electret Media for HVAC Filtration Applications," The AFS Indoor Air
Quality-Filtration Conference Proceedings, Cincinnati, OH, November 14-15, 2002.
Barrett, L. W. and A.D. Rousseau. "Aerosol Loading Performance of Electret Filter Media," American
Industrial Hygiene Association Journal 59: 532-539 (1998).
Barris, M.A., M.A. Gogins, and T.M. Weik. "Filtration Arrangement Utilizing Pleated Construction
and Method," Donaldson Company, Inc., U.S. Patent 6,800,117, October, 2004.
Benson, J.D., D.G. Crofoot, M.A. Gogins, and T.M. Weik. "Filter Structure with Two or More
Layers of Fine Fiber Having Extended Useful Service Life," Donaldson Company, Inc., U.S. Patent
6,746,517, June 2004.
Bergman, W., A. Biermann, W. Kuhl, B. Lum, A. Bogdanoff, H. Hebard, M. Hall, D. Banks, M.
Mazumderm, and J. Johnson. "Electric Air Filtration: Theory, Laboratory Studies, Hardware
Development, and Field Evaluations," Technical Report, NTIS, PC A13/MF A01, Lawrence
Livermore National Lab., September 1983.
Chung, H.Y., J.R.B. Hall, M.A. Gogins, D.G. Crofoot, and T.M. Weik. "Polymer, Polymer Microfiber,
Polymer Nanofiber and Applications Including Filter Structures," Donaldson Company, Inc., U.S.
Patent 6,743,273, June 2004.
Edward, M.E. Jr. "Use of Biological Weapons." Chapter 20, Medical Aspects of Chemical and
Biological Warfare, The Textbooks of Military Medicine, Borden Institute, Walter Reed Army Medical
Center, Washington, D.C., 1997.
Gillingham, G.R., M.A. Gogins, and T.M. Weik. "Air Filtration Arrangements Having Fluted Media
Constructions and Methods," Donaldson Company, Inc., U.S. Patent 6,673,136, January 2004.
Gogins, M. A., and T.M. Weik. "Air Filter Assembly for Filtering an Air Stream to Remove
Paniculate Matter Entrained in the Stream," Donaldson Company, Inc., U.S. Patent 6,716,274,
April 2004.
Graham, K., M. Ouyang, T. Raether, T Grafe, B. McDonald, and P. Knauf, P. "Polymeric Nanofibers
in Air Filtration Applications," 15th Annual Technical Conference & Expo of the American Filtration
& Separations Society, Galveston, Texas, April 2002.
Hanley, J.T., D.S. Ensor, K.K. Foarde, L.E. Sparks. "The Effect of Loading Dust Type on the
Filtration Efficiency of Electrostatically Charged Filters," Indoor Air 99, Edinburgh, Scotland,
August, 1999.
Hanley, J.T. and M.K. Owen. "Develop a New Loading Dust and Dust Loading Procedures for the
ASHRAE Filter Test Standards 52.1 and 52.2, Final Report," ASHRAE Project No. 1190-RP,
August 2003.
-------
Homonoff, E. "North American Filtration Markets: 2003-2008," Filtration 2004 International
Conference and Exposition in Philadelphia, December 7-9, 2004.
Jaisinghani, R.A., TJ. Inzana, and G. Glindemann. "New Bactericidal Electrically Enhanced
Filtration System for Cleanrooms," IEST 44th Annual Technical Meeting, Phoenix, AZ, April, 1998.
Janssen, L.L., J.O. Bidwell, H.E. Mullins, and TJ. Nelson. "Efficiency of Degraded Electret Filters:
Part I - Laboratory Testing Against NaCl and DOP Before and After Exposure to Workplace
Aerosols," Journal of the International Society for Respiratory Protection 20: 71-80
(Fall/Winter 2003a).
Janssen, L.L., J.O. Bidwell, H.E. Mullins, and TJ. Nelson. "Efficiency of Degraded Electret Filters:
Part II - Field Testing Against Workplace Aerosols," Journal of the International Society for
Respiratory Protection 20: 81-90 (Fall/Winter 2003b).
Lehtimaki, M. and K. Heinonen. "Reliability of Electret Filters," Building and Environment
29(3): 353-355(1994).
Lehtimaki, M. "Development of Test Methods for Electret Filters," Nordtest, NT Technical Report
320. 57p. NT Project No. 1164-94, 1996.
Lifshutz, N. "Performance Decay in Synthetic Electret Filter Media," Advances in Filtration and
Separation Technology 11: 307-311 (1997).
Myers, D.L. and B.D. Arnold. "Electret Media for HVAC Filtration Applications," Paper presented
at INTC 2003, International Nonwovens Technical Conference, Conference Proceedings,
pp. 602-630 (2003).
Pierce, M.E. and N. Lifshutz. "New Developments in Synthetic ASHARE Filtration Media," TAPPI
Journal 80: 142-145 (1997).
Raynor, PC. and S.J. Chae. "Effects of Particle Loading on Electrostatically Charged Filters in an
HVAC System," the AFS Indoor Air Quality-Filtration Conference Proceedings, Cincinnati, OH,
November 14-15, 2002.
Raynor, PC. and S.J. Chae. "Dust Loading on Electrostatically Charged Filters in a Standard Test and
a Real HVAC System," Filtration & Separation, pp. 35-39 (March 2003).
Romay, F.J. and B.Y.H. Liu. "Degradation of Electret Filters During DOP Aerosol Loading,"
Advances in Filtration and Separation Technology, V 12, Advancing Filtration Solutions, St. Louis,
Missouri. American Filtration & Separations Society, pp. 193-200, (May 4-7, 1998).
Romay, F.J., B.Y.H. Liu, and S.J. Chae. "Experimental Study of Electrostatic Capture Mechanisms in
Commercial Electret Filters," Aerosol Science and Technology 28: 224-234 (1998).
-------
Appendix A
Descriptions of the Databases Searched
Table A-l. Descriptions of the Databases Searched
Database
CBIAC
Producer
Chemical Warfare/Chemical
and Biological Defense
Information Analysis
Center (CBIAC)
File Size
More than 1.4
million records
Content/Description
The CBIAC, operated by Battelle. is a full-service Department
of Defense (DoD) Information Analysis Center (IAC). The
CBIAC maintains a database containing more than 103,000
document citations, as well as an on-site collection of more
than 38,000 books, technical reports, videotapes, and
magnetic diskettes from domestic and foreign sources.
DTIC
The Defense Technical
Information Center
More than 2
million records
The DTIC databases contain technical reports, patents, journal
articles, conference proceedings, thesis related to defense-
sponsored research, development, test, and evaluation efforts.
CA Search
Chemical Abstracts Service
(CAS)
More than 22.2
million records
The CA Search database covers all areas of biochemistry,
chemistry, and chemical engineering. It contains records
for documents reported in printed Chemical Abstracts (CA).
The records come from the 1,300 core journals and patents
from 26 countries and 2 international patent organizations.
Technical reports, books, conference proceedings, and
dissertations are also included.
NTIS
National Technical
Information Service
More than 2.2
million records
The National Technical Information Service database
contains abstracts on government-sponsored research, which
corresponds to Government Reports Announcement & Index.
The file contains records for all areas of science, engineering,
and technology. The sources are publications on research,
development, and engineering projects sponsored by U.S. and
other governments.
Energy SciTec
Office of Scientific and
Technical Information, U.S.
Department of Energy
More than 4.4
million records
The Energy Science & Technology database covers worldwide
literature on energy research and technology for all kinds of
energy sources, including environmental and other related
aspects. Citations in the database are from journals, series,
reports, conference papers, books, and patents.
Ei Compendex
Elsevier Engineering
Information, Inc.
More than 4.6
million records
The Ei Compendex® database is the machine-readable
version of the Engineering Index, which provides abstracted
information from the world's significant engineering and
technological literature. The Compendex database provides
worldwide coverage of approximately 4,500 journals and
selected government reports and books. The database covers
all engineering disciplines, including chemical, energy,
environmental, biological engineering, etc.
SciSearch
Institute for Scientific
Information (ISI)
More than 12.1
million records
The SciSearch®, a cited reference science database, is an
international, multidisciplinary index to the literature of
science, technology, biomedicine, and related disciplines.
SciSearch contains all of the records published in the
Science Citation Index® (SCI®), plus additional records
in engineering technology, physical sciences, agriculture,
biology, environmental sciences, clinical medicine, and the
life sciences. SciSearch indexes all significant items (articles,
review papers, meeting abstracts, letters, editorials, book
reviews, correction notices, etc.) from more than 6,100
international scientific and technical journals.
Biosis Previews
BIOSIS
More than 20
million records
The BIOSIS Previews® database contains citations from
Biological Abstract^1 (BA), and Biological Abstracts/Reports,
Reviews, and Meetings® (BA/RRM). The database provides
comprehensive worldwide coverage of research in the
biological and biomedical sciences.
-------
Table A-l. Descriptions of the Databases Searched (Continued)
Database
Enviroline
Producer
Congressional Information
Service, Inc.
File Size Content/Description
More than 0.3 The Enviroline® database covers the world's environmental
million records related information. It provides indexing and abstracting
coverage of more than 1,000 international primary and
secondary publications reporting on all aspects of the
environment. Enviroline corresponds to the print Environment
Abstracts.
World Textiles
Elsevier
More than 0.3
million records
The World Textiles™ database covers the worldwide literature
on the science and technology of textiles and related
materials. The database, which includes the coverage of World
Textile Abstracts, offers comprehensive coverage of the world's
textile-related literature from technical, scientific, economic,
and commercial journals, and statistical publications. In
addition, World Textiles™ includes unique coverage of
the related patents and patent applications from the US,
European, and British patent offices.
Textile Technology
Digest
Institute of Textile
Technology
More than 0.3
million records
The Textile Technology Digest database provides international
coverage of the literature of textiles and related subjects.
Coverage includes the various aspects of textile production
and processing. Textile Technology Digest corresponds to the
print publication of the same name.
-------
-------
&EPA
United States
Environmental Protection
Agency
PRESORTED STANDARD
POSTAGES FEES PAID
EPA
PERMIT NO. G-35
Office of Research and Development
National Homeland Security Research Center
Cincinnati, OH 45268
Official Business
Penalty for Private Use
$300
Recycled/Recyclable
Printed with vegetable-based ink on
paper that contains a minimum of
50% post-consumer fiber content
processed chlorine free
------- |