Performance Evaluation of
Air Filtration Devices
FINAL REPORT

Prepared for the
South Coast Air Quality Management District
21865 Copley Dr, Diamond Bar, CA 91765

by

Principal Investigator: Dr. Robert L. Russell
Co-Principal Investigators: William A. Welch & James Gutierrez
College of Engineering-Center for Environmental Research and Technology
University of California, Riverside
1084 Columbia Avenue
Riverside, CA 92507
Tel: (951) 781-5791
Fax: (951) 781-5790
i

 DISCLAIMER
The statements and conclusions in this report are those of the contractor and not
necessarily those of the South Coast Air Quality Management District (AQMD). The mention of
commercial products, their source, or their use in connection with material reported herein is not
to be construed as actual or implied endorsement of such products.

ACKNOWLEDGEMENTS
This report was prepared at the University of California, Riverside, Bourns College of
Engineering-Center for Environmental Research and Technology (CE-CERT). The author would
like to thank the following organization and individuals for their valuable contributions to this
project: AQMD for providing the instruments to measure the targeted air pollutants; Dan
Baldwin, Director of Risk Management, Facilities, and Fleet Services, Jurupa Unified School
District, and Gary Dixon, Principal of Sunnyslope Elementary School, for allowing us access to
the school site and a classroom in which to perform the measurements. Funding for this work
was provided by AQMD.

 Draft Report: Performance Evaluation of Air Filtration Devices

TABLE OF CONTENTS
DISCLAIMER................................................................................................................................ i 
ACKNOWLEDGEMENTS .......................................................................................................... i 
TABLE OF CONTENTS ............................................................................................................. ii 
EXECUTIVE SUMMARY .......................................................................................................... 3 
INTRODUCTION......................................................................................................................... 5 
EXPERIMENTAL PROCEDURES ........................................................................................... 6 
Testing Location ......................................................................................................................... 6 
Measurement Equipment Setup and Testing Procedure ............................................................. 7 
Baseline Measurements ............................................................................................................ 10 
Additional Measurements ......................................................................................................... 10 
HVAC Specifications................................................................................................................ 12 
Devices Tested .......................................................................................................................... 12 
Relative and Overall Removal Efficiency ................................................................................ 14 
Performance Criteria ................................................................................................................. 15 
TEST RESULTS ......................................................................................................................... 15 
Determination of the Air Exchange Rate .................................................................................. 15 
Removal Efficiency .................................................................................................................. 16 
Sound Level Measurements ...................................................................................................... 20 
Ozone Measurements ................................................................................................................ 21 
REFERENCES............................................................................................................................ 22 
APPENDIX .................................................................................................................................. 23 

 
 

ii

 EXECUTIVE SUMMARY
The goal of this testing program was to evaluate the ability of various air filtration
devices to reduce the outdoor infiltrated concentrations of ultrafine particles (UFP; particles with
an aerodynamic diameter equal or less than 100 nm) and black carbon (BC; an indicator of diesel
particles) in a typical classroom setting. Over 150 manufacturers were contacted by South Coast
Air Quality Management District (AQMD) staff, and nine companies participated in this
program with a total 15 different air filtration devices submitted for testing. Specifically, ten
panel filters, four stand-alone units, and one register system were submitted. All measurements
were conducted between 07/08/10 and 08/03/10 by the University of California Riverside Center
for Environmental Research & Technology (CE-CERT) in a portable classroom at Sunnyslope
Elementary School in Riverside, CA. The use of this type of prefabricated structure is becoming
more and more prevalent in United States public schools, and it has been estimated that in
California, around one third of the students spend time in portable classrooms during a typical
school day (Shendell et al. 2004).
For this program, UFP and BC were measured at different distances from the device
tested and both indoors and outdoors, and each technology was examined for a minimum of six
consecutive hours. Baseline measurements were taken before installing any of the air purification
solutions to estimate the pre-existing relative and overall removal efficiencies of the classroom
before modification. Measurements also included: (1) the pressure drop across the panel filters in
the Heating, Ventilation, and Air Conditioning (HVAC) system, (2) the noise level and ozone
generation of the stand-alone devices, and (3) the air exchange rate (AER) of the room.
For the purpose of this study, the performance of each air filtration device was estimated
by measuring the relative and the overall particle removal efficiencies, which were defined as
follows:
Relative Removal Efficiency: percentage reduction in the concentration of UFP (or BC)
downstream of the device relative to its concurrent ambient
(outdoor) level upstream of the HVAC, demonstrated during the six
hour testing period
Overall Removal Efficiency: percentage reduction in the concentration of UFP (or BC) in the
center of the classroom and at breathing height relative to its
concurrent ambient (outdoor level), also demonstrated during the six
hour testing period
Only air filtration devices that satisfied the following minimum pre-specified
requirements were accepted as approved technologies:
•
•
•
•
•

Overall removal efficiencies for UFP and BC of at least 85% (for panel filters and register
systems)
Relative removal efficiency for UFP and BC of at least 85% (for stand-alone units)
Low pressure drop across the filter (for panel filters)
No ozone generation (<5 ppb)
A noise level below 45 decibels [db(A)] (for stand-alone units)

3

 Draft Report: Performance Evaluation of Air Filtration Devices
The main results of this testing program have been summarized below:
•

•

•

For panel filters, the overall (and relative) particle removal efficiencies increased with
increasing Minimum Efficiency Reporting Value (MERV) rating. The “Nanomax S-220”
manufactured by IQAir (a MERV 16 panel filter) was the only HVAC-mounted device that
satisfied the performance requirements set by AQMD, and its overall removal efficiencies
were between 88 and 91% for UFP and BC, respectively.
All stand-alone units appeared to have a substantially higher removal efficiency immediately
downwind of the device (relative removal efficiency) than at breathing level (overall removal
efficiency). This was due to the distance between the indoor real-time monitors and the air
supply registers, the lower air flow rate “processed” by these air filtration devices relative to
the flow rate supplied by the HVAC system, and other factors intrinsic to our particular
classroom set-up. Therefore, relative removal efficiency was the most robust and appropriate
measurement used to evaluate the performance of these types of devices. In this respect, all
stand-alone units tested showed high removal efficiencies with values varying from 94 to
100% and between 83 and 94% for UFP and BC, respectively. However, only the
“CleanZone SL” manufactured by IQAir did not exceed a noise level of 45 decibels [dB(A)],
a threshold set by many school districts for new in-classroom equipment.
All stand-alone units and the register mounted system (which employs an electrostatic filter)
were found to generate no measureable ozone levels.

4

 Draft Report: Performance Evaluation of Air Filtration Devices

INTRODUCTION
In 2009, the South Coast Air Quality Management District (AQMD) conducted a pilot
study to investigate the effectiveness of different air purification systems/solutions in reducing
the exposure of children to outdoor-infiltrated and indoor-generated air contaminants inside nine
classrooms at three Southern California schools (AQMD, 2009). The introduction in their report
notes that numerous epidemiological and toxicological studies have found positive associations
between exposure to atmospheric particulate matter (PM) and adverse health effects (Pope and
Dockery, 2006; Environmental Protection Agency Integrated Science Assessments, 2009).
Although air quality standards have been established for outdoor ambient environments, a
significant portion of human exposures to PM occurs indoors, where people spend around 8590% of their time. Hence, it is important to understand and reduce the sources of both indoor and
outdoor PM. Indoor PM consists of outdoor particles that have infiltrated indoors, particles
emitted indoors (primary), and particles formed indoors (secondary) from precursors emitted
both indoors and outdoors. Children are regarded as particularly susceptible to potential health
hazards related to PM exposure, which includes asthma, lung inflammation, allergies and other
types of respiratory and cardiovascular problems. School-aged children spend approximately
30% of their day in classrooms. For this reason, minimizing the concentration of PM (as well as
that of other air contaminants) inside classrooms is important, especially at schools located in
close proximity to roadways and other substantial sources of air pollution.
Earlier in 2010 AQMD released an Announcement of a Testing Opportunity of Air
Filtration Technologies (PON2010-02) to solicit a list of qualified air filtration devices (panel
filters, register-based air purifiers, and stand-alone units) to be used in the installation and
maintenance of air filtration systems in Wilmington area schools using $5.4 million in funds
from the TraPac Settlement Special Revenue Fund. In accordance with AQMD’s Procurement
Policy and Procedure, notice of this testing opportunity was published in the Los Angeles Times,
the Orange County Register, the San Bernardino Sun, and Riverside County Press Enterprise
newspapers to leverage the most cost-effective method of outreach to the entire South Coast Air
Basin. Additionally, potential vendors were notified utilizing AQMD’s TAO mailing list and
other mailing lists as appropriate and by conducting outreach at industry conferences.
To qualify for free testing of air filtration technologies for removal of ultrafine particles
(UFP; particles with an aerodynamic diameter equal or less than 100 nm) and black carbon (BC;
an indicator of diesel particles), 20” x 30” x 1” (or 2” maximum) deep panel filters and/or
register-based air purifiers and/or stand-alone units were submitted by various US manufacturers
to the University of California Riverside College of Engineering Center for Environmental
Research & Technology (CE-CERT) between 05/07/10 and 06/30/10. Each manufacturer was
allowed to submit no more than three air filtration technologies for testing. The specific goal of
this program was to evaluate the efficiency of all received air filtration devices for reducing the
outdoor-infiltrated and indoor-generated concentrations of UFP and BC in a typical classroom
setting. All measurements were made in a portable classroom at Sunnyslope Elementary School
in Riverside, California, from 07/08/10 through 08/03/10. Recently, the use of prefabricated,
portable classrooms has increased in United States public schools, and in California
approximately one of three students spends time inside this type of structure during a typical
school day (Shendell et al. 2004).
Additional measurements conducted during this testing program also included: (1) the
pressure drop across the panel filters in the Heating, Ventilation, and Air Conditioning (HVAC)
5

 Draft Report: Performance Evaluation of Air Filtration Devices
system, (2) the noise level and ozone generation of the stand-alone devices, and (3) the air
exchange rate (AER) of the room.
Over 150 manufacturers were contacted by AQMD staff and nine manufacturers
submitted a total of 15 different air filtration devices for evaluation (i.e. ten panel filters, four
stand-alone units, and one register system). This report describes the procedure designed to
assess the performance of all submitted air filtration systems and the main results of this testing
program.

EXPERIMENTAL PROCEDURES
Testing Location
The performance evaluation of all submitted panel filters, stand-alone units, and of a
register mounted system was conducted in a portable classroom located at Sunnyslope
Elementary School, 7050 38th St, Riverside, CA (Figure 1) that was specifically selected for the
purpose of this testing. This school is located within 1000 feet north of a major freeway (Pomona
freeway) and the average fine particulate matter (PM2.5; particles with an aerodynamic diameter
equal or less than 2.5 µm) concentration in this area is generally high. Pictures of the outside of
the classroom are shown in the Appendix (Figures A-1 and A-21).

Figure 1 Aerial view of Sunnyslope Elementary School and of the surrounding area
The classroom is 39 feet long, 23 feet wide, and 8.5 feet high to the false ceiling (~10.5
feet to the permanent ceiling), and it is conditioned by an HVAC manufactured by BARD Inc.
(model WH483-A10xx4xxx heat pump; Figure A-3). The setup of the equipment took place
between 7/7/10 and 7/8/10 and all measurements were conducted on weekdays from 7/8/10
1

See Appendix

6

 Draft Report: Performance Evaluation of Air Filtration Devices
through 8/3/10. No children were present during testing. The measurement equipment was
removed on 8/4/10.
Measurement Equipment Setup and Testing Procedure
Ten panel filters, four stand-alone units, and one register system were installed inside the
selected classroom (one at a time) according to the manufacturer specifications, and tested for
their ability to reduce the indoor concentrations of UFP and BC. For this purpose, UFP and BC
were measured at up to four positions using real-time analyzers connected to a central data
logger (Figure 2):
Position #1:

Outside the classroom and immediately upwind of the HVAC system intake
(when testing panel filters, register systems, and stand-alone units)

Position #2:

In the center of the classroom at about three feet from the floor (the
approximate height of a child’s head when seated), away from all registers, and
just a few meters from the student area (when testing panel filters, register
systems, and stand-alone units)

Position #3:

a. Upstream of the return air intake duct in the classroom (when testing panel
or register filters)
b. As close to the air intake of the stand-alone as possible, with the HVAC
system on, and with no filtration device mounted on the HVAC register (when
testing stand-alone units)

Position #4:

a. As close to the downstream side of the filter or register systems as
possible (when testing panel filters or register systems)
b. As close to the outlet of the stand-alone as possible (when testing standalone units)

7

 Draft Report: Performance Evaluation of Air Filtration Devices

HVAC

a)
Register #1

Register #2

#4
#3
Return Air

~ 3 feet

#1
#2

#1: Ambient / upstream of HVAC
#3: Upstream of return air duct
Panel Filter

#2: Classroom
#4: Downstream of filter

b)
Register System

#4
HVAC

#3
Return Air

#1
#2
~ 3 feet

#1: Ambient / upstream of HVAC
#3: Upstream of return air duct
Panel Filter

#2: Classroom
#4: Downstream of register

8

 Draft Report: Performance Evaluation of Air Filtration Devices

Register #1

Register #2

#4

#3
#1: Outside the classroom
#3: Stand-Alone Inlet

~ 3 feet

STAND
ALONE

Return Air

HVAC

c)

#1

#2

#2: Inside the classroom
#4: Stand-Alone outlet

Figure 2 Configuration used for testing the particle removal efficiency of a) panel filters, b)
register systems, and c) stand-alone units. The approximate positions of the four mobile air
monitoring stations that were used to measure the indoor (in-classroom) and ambient (outdoor)
concentrations of UFP and BC are shown as numbers 1 through 4
It should be noted that one of the two outlet registers was properly sealed prior to testing
any register systems (Figure 2b). Sampling at all positions was conducted in the morning / early
afternoon for a minimum of six consecutive hours during which the HVAC system was operated
at the normally used set point. All test measurements were conducted under the same repeatable
conditions, with the HVAC system on and when the room was not occupied by students. When
testing all air filtration devices the HVAC unit was set to control the room temperature to 74 ºF
and the fan was run continuously.
It is worth noting that while the standard MERV rating system for filters may adequately
address PM2.5 mass removal efficiency, these ratings do not address particles below 300 nm and
fresh diesel PM which is mostly less than 200 nm. Thus, air filtration technologies are not
typically tested for removal of smaller particles such as UFP and those particles containing most
of the BC. Given the potential challenges of filtering out these smaller particles combined with
evidence associating them with increased toxicity or cancer risk, the main focus of our testing
program was on measuring the ability of commercial air filtration systems to remove UFP and
BC from the indoor air of classrooms. For this purpose, four mobile air quality monitoring
stations were used to measure the indoor and outdoor concentrations of these two targeted air
pollutants (Figures A-4 and A-5). Each of these stations was comprised of a small table
supporting the following instruments:
•

A portable Aethalometer (model AE42, Magee Scientific, 2800 Adeline St., Berkeley CA
94703) to provide continuous BC concentration measurements (ng/m3) at five minute
intervals. BC is a component of PM indicative of diesel and soot particles from combustion
processes, and typically is found in smaller sized particles.
9

 Draft Report: Performance Evaluation of Air Filtration Devices
•

A water-based condensation particle counter (CPC model 3781, TSI, 500 Cardigan Road,
Shoreview, MN 55126) to provide continuous measurements of the particle number
concentration (#/cm3; an indicator of UFP) at one minute intervals. This CPC model
measures the number of particles down to at least 10 nm in diameter.

Before and after testing each air purification device, the four measurement stations were
collocated inside the test class-room and all instruments run “side-by-side” for 30 minutes to
provide quality assurance of the measurements, to estimate the precision characteristics, and to
identify any potential problems. The onsite instrument operators maintained a written log-book
of any conditions which may have affected the measurements. They also monitored the results as
they were displayed on a computer screen and noted in the log-book if they observed any
readings which appeared to be abnormal. At the end of each day of testing, the data were
downloaded to a disc or a flash drive and given to the CE-CERT principal investigator (PI) for
an initial evaluation.
Baseline Measurements
Baseline measurements were taken before installing any of the air purification solutions
to estimate the pre-existing relative and overall removal efficiencies of the classroom before
modification. For this purpose a clean “low efficiency” panel filter (of the same type and brand
usually employed by Sunnyslope Elementary School; Figures A-6 through A-8) was mounted
inside the HVAC system and UFP and BC measured at positions 1 through 4 as described in the
previous section (Figure 2a). In this case, sampling was conducted in the morning / early
afternoon for five selected days from 07/08/10 to 07/14/10 and for a minimum of six hours per
day during which the HVAC was operated at the normally used set point with the fan set to stay
on. To ensure a more direct comparison between all devices tested, baseline conditions for standalone devices were established with the HVAC turned on but with no filtration device mounted
on the HVAC register. All baseline measurements were conducted under repeatable conditions
when the room was not occupied by students.
Additional Measurements
A 0 to 1” water Magnehelic gauge was used to monitor the pressure drop across the
HVAC filter. When evaluating the performance of air filtration devices using components that
might generate ozone (e.g. air ionizer purifiers), an ozone detector (Dasibi model 1003AH) was
used to measure any potential variations in the indoor concentration of this pollutant. Also, the
noise level produced by the stand-alone units was measured at several locations in the room with
an Extech 403407 decibel meter. The measurements were made with the HVAC off and with the
meter pointed at the stand-alone unit while held at ~3 feet above the floor at the locations
indicated in Figure 3. The distance from the stand-alone to the numbered locations are: 1 = 40”,
2 = 10’ 40”, 3 = 20’ 40”, 4 = 20’ 40” but 5’ north of position 3, 5 = 20’ 40” but 7’ south of
position 3, and 6 = 30’ 40”.

10

 Draft Report: Performance Evaluation of Air Filtration Devices

Window
HVAC
5

Stand-Alone

23 feet

Register System

1

2

3

6
Window

4

Entry Door

39 feet
Figure 3 Schematic of the test classroom showing where the noise measurements were taken
The indoor-outdoor air exchange rate (AER; hr-1) of the test room was estimated by
injecting a known amount of carbon monoxide (CO) into the HVAC ambient air intake and
monitoring the decay with a commercial IR analyzer (Horiba PG250 multi-gas analyzer).
Specifically, a Teflon tube connected to a cylinder containing 15% hydrogen (H2) in 85% CO
was placed into the grill where ambient air enters the HVAC. The valve of the cylinder was then
opened and the H2/CO mixture flowed through a flow meter until at least 50 ppmv of CO had
been injected into the room. No other significant source of CO was present inside the classroom
when these measurements were taken. Assuming an exponential decay of particles, that AER and
outdoor concentrations are constant during the decay period, and that indoor concentrations are
well mixed, then:
(1)
or,
(2)
where, Ct is the indoor CO concentration after time t (after the decay period), C0 is the initial
peak CO concentration (right after CO emission), and k is the indoor loss rate for particles or
gases (hr-1; Abt et al. 2000). Because k is rather negligible for CO, it was possible to estimate the
AER for the test classroom directly from eq (2) by regressing lnCt over t.

11

 Draft Report: Performance Evaluation of Air Filtration Devices

HVAC Specifications
While BARD has discontinued the production of the WH48 heat pump used in the test
classroom, they have maintained all of the specification information for this and other
discontinued models on their website. This state that the rated air flow rate for this particular
model is 1550 cfm and that the total air flow at high speed will decrease from 1885 cfm at 0”
water static pressure across the high speed tap to 1285 cfm at 0.4” water static pressure at the
high speed tap. The ventilation flow (i.e. the flow of ambient air being mixed with the room
return air flow) increases from ~10% at 0” water static pressure to ~25% at 0.4” water static
pressure.
Devices Tested
Nine different manufacturers participated in the testing program and a total of 15
different air filtration technologies were evaluated. Three of the submitted devices were not
tested because of their incompatibility with the HVAC system or due to technical difficulties
related to their installation. In particular, one of the panel filters submitted by Camil Farr was too
large to be mounted in the BARD panel filter holder. The Genesis Air Photo-catalytic Panel,
designed to be connected inside the solid aluminum ductwork employed in permanent buildings,
did not fit inside the corrugated ductwork present in the portable classroom used for these tests.
Environmental Dynamics Group submitted an electrostatic HVAC-based panel filter for which
performance could not be evaluated because the unit was not supplied with the transformer
required for its proper operation.
A list of all air filtration technologies submitted to and tested by CE-CERT along with
the manufacturers’ names, model numbers, device dimensions, MERV rating and pressure drop
across the filter (for panel filers only), and the corresponding IDs used throughout this report can
be found in Table 1.

12

 Draft Report: Performance Evaluation of Air Filtration Devices
Table 1 Information about all devices tested including manufacturers’ names, model numbers,
filter dimensions, MERV rating and pressure drop across the filter (for panel filers only), and the
corresponding IDs used throughout this report
Manufacturer

Model

Nominal
Size

MERV
rating

Delta P
(inches water)

ID in
report

Existing Filter
(Baseline)

NA

20” x 30” x 1”

8

0.14

EF-8

20” x 30” x 1”

8

0.10

CF-8

20” x 30” x 1”

7

Not recorded

WE1-7

20” x 30” x 2”

7

0.14

WE2-7

Ultra 1500

20” x 30” x 1”

11

0.13

U-11

Maxx 2000

20” x 30” x 1”

12

0.14

M-12

Exceed

20” x 30” x 2”

11

0.10

E-11

Exceed

20” x 30” x 2”

13

0.12

E-13

Exceed

20” x 30” x 2”

14

0.24

E-14

20” x 30” x 2”

13

0.08

IQ-13

20” x 30” x 2”

16

0.13

IQ-16

72” x 29” x 10”

500

NM

CZ

13” x 13” x
21.5”
32” x 26” x 13”

140 - 3502

NM

PA

3

195 - 545

4

1.1

NQ-1

28” x 15” x 14”

3505

NM

NQ-2

51” x 57” x 14”

NA

NM

EDG

Camil Farr
Air Cleaners, Inc.
Air Cleaners, Inc.
Freudenberg
Filtration
Technologies L. P.
Freudenberg
Filtration
Technologies L. P.
eSpin
Technologies, Inc.
eSpin
Technologies, Inc.
eSpin
Technologies, Inc.
IQ Air
IQ Air
IQ Air
Pure Air
NQ Industries Inc.
NQ Industries Inc.
Environmental
Dynamics Group,
Inc.

®

30/30 –
M8
Washable
Electrostatic
Washable
Electrostatic

IQ Air
MERV 13
Nanomax S220
CleanZone
SL
HPS 350
NQ400
NQ
Clarifier
1V8-4812291/2

NA = Not applicable
NM = Not measured
2
Four switchable flow rates between 140 and 350 cfm; tested at 350 cfm
3
Four switchable flow rates between 195 and 545 cfm; tested at 545 cfm. Also contains one UV lamp
4
Reading on gage of NQ-1
5
Variable speed fan up to a maximum flow rate of 350 cfm; tested at 350 cfm. Also contains two UV lamps

13

 Draft Report: Performance Evaluation of Air Filtration Devices

Relative and Overall Removal Efficiency
The Relative Removal Efficiency of each air filtration device tested was estimated as:
a)

The percentage reduction in the concentration of UFP (or BC) downstream of the
device tested (position #4; Figure 2) relative to its concurrent ambient (outdoor)
level upstream of the HVAC (position #1), demonstrated during the six hour
testing period.
 

100

#  

 

#

/

(3)

#

Similarly, the Overall Removal Efficiency for each air filtration device tested was defined
as the percentage reduction in the concentration of UFP (or BC) inside the classroom (position
#2; Figure 2) relative to its concurrent ambient (outdoor) level (position #1), demonstrated
during the six hour testing period.
100

#  

 

#

/

(4)

#

where,
P#1 = particle number or BC concentration at position #1 (ambient) expressed in #/cm3
and μg/m3, respectively
P#4 = particle number or BC concentration at position #4 (downstream of the device)
expressed in #/cm3 and μg/m3, respectively
P#2 = particle number or BC concentration in position #2 (inside the classroom and at
breathing height) expressed in #/cm3 and μg/m3, respectively
Measurement days affected by exceptional meteorological conditions (e.g. rain) or
instrument malfunction were discarded from these calculations.
Condensation Particle Counters were set to store the date and time, error flags, and
particle concentration after every minute. After combining the four individual CPC
measurements (positions 1 through 4; Figure 2) into a single file aligned by date and time, the
relative and overall removal efficiencies for each device were calculated on a minute by minute
basis and then averaged over the entire testing period. Similarly, Aethalometers were set to
record the date and time, BC concentration, error flags, and other data after every 5 minutes. All
data were analyzed and smoothed using the WUAQL AETHALOMETER DATA MASHER
(Version 6.0h, May 22, 2008; Air Quality Laboratory at Washington University in St. Louis),
and the four individual Aethalometer measurements (positions 1 through 4; Figure 2) were
combined into a single file aligned by date and time. Also in this case, the five minute average
removal efficiencies for BC were averaged over the entire testing interval.

14

 Draft Report: Performance Evaluation of Air Filtration Devices

Performance Criteria
This testing procedure was designed to select the air filtration devices that provide the
most substantial improvement in air quality with respect to baseline conditions. The best
performing technologies are the ones that satisfy the following specifications set by AQMD:
•

•
•
•

High removal efficiency
o For panel filters and register systems: both of these technologies are installed in
line with the HVAC conditioning system, require forced air to work properly, and
are characterized by a similar principal of operation. An overall removal
efficiency for UFP and BC of at least 85% was set as the lowest acceptable
performance requirement for all panel filters and register systems tested
o For stand-alone systems: as described in more detail in the results section, the
overall removal efficiency for stand-alone devices was affected by their relative
distance to the air supply registers and by other factors intrinsic to the particular
classroom set-up. Therefore, in this case the relative removal efficiency was
found to be a better indicator of performance than the overall removal efficiency.
A relative removal efficiency for UFP and BC of at least 85% was set as the
lowest acceptable performance requirement for all stand-alone units tested
Low pressure drop across the panel filters
No ozone generation (<5 ppb)
A noise level below 45 decibels [db(A)] to conform with the noise threshold set by many
school districts for new in classroom equipment (for stand-alone units)

TEST RESULTS
Determination of the Air Exchange Rate
Measurements of the AER were conducted on 8/3/10. At that time a cylinder containing
15% hydrogen / 85% carbon monoxide with a flow meter attached to it was placed near the
outside of the HVAC. A Teflon tube from the flow meter was inserted into the HVAC ambient
air inlet grill and approximately 15 to 20 liters of the gas were injected into the HVAC to
produce a CO concentration in the room of approximately 70 ppmv. Figure 4a shows the buildup
and decay of CO in the room as recorded every second for about 1.5 hr by the Horiba PG250.
The AER was then estimated from eq (2) by regressing lnCt over t, and it was found to be 2.1 hr1
(Figure 4b). Only data from the moment CO peaked to the maximum concentration (after 10.3
minutes from the initial injection) to the end of the collection period (slightly before all CO
exited the classroom) were considered in our calculations.
These results suggest that the AER of the test room in Sunnyslope was within the typical
values reported for similar portable classrooms in other parts of Los Angeles County (Shendell et
al. 2004), where ventilation rates were found to range between 0.1 and 2.9 hr-1.

15

 Draft Report: Performance Evaluation of Air Filtration Devices
a)

b)

Figure 4 Buildup and decay of CO in the room as recorded by the Horiba PG250 (a). The AER
was estimated from eq (2) by regressing lnCt (ln-transformed concentration of CO) over t (time)

Removal Efficiency
Figure 5 shows relative and overall UFP removal efficiencies for the register system and
all panel filters (a), and for all stand-alone units (b) tested; results are compared to the
corresponding baseline conditions. Similarly, Figure 6 illustrates relative and overall BC removal
efficiencies for the register system and all panel filters (a), and for all stand-alone units (b)
tested; also in this case, results are compared to the corresponding baseline conditions. Removal
efficiency data have also been summarized in Tables 2 and 3.

16

 Draft Report: Performance Evaluation of Air Filtration Devices

a)

Relative Removal Efficiency

Overall Removal Efficiency

Removal Efficiency (%)

100
90
80

ULTRAFINE
PARTICLES

70
60
50
40
30
20
10

b)

EDG

IQ-16

IQ-13

E-14

E-13

E-11

M-12

U-11

CF-8

WE2-7

WE1-7

BASELINE

0

Relative Removal Efficiency

Removal Efficiency (%)

100
90
80

ULTRAFINE
PARTICLES

70
60
50
40
30
20
10

PA

NQ-2

NQ-1

CZ

BASELINE

0

Figure 5 Relative and overall UFP removal efficiencies for the register system and all panel
filters (a), and for all stand-alone units (b) tested. Corresponding baseline conditions are also
included

17

 Draft Report: Performance Evaluation of Air Filtration Devices

a)

Relative Removal Efficiency

Overall Removal Efficiency

Removal Efficiency (%)

100
90
80

BLACK
CARBON

70
60
50
40
30
20
10

EDG

IQ-16

IQ-13

E-14

E-13

E-11

M-12

U-11

CF-8

WE2-7

Relative Removal Efficiency

PA

NQ-2

NQ-1

BLACK
CARBON

CZ

100
90
80
70
60
50
40
30
20
10
0

BASELINE

Removal Efficiency (%)

b)

WE1-7

BASELINE

0

Figure 6 Relative and overall BC removal efficiencies for the register system and all panel filters
(a), and for all stand-alone units (b) tested. Corresponding baseline conditions are also included

18

 Draft Report: Performance Evaluation of Air Filtration Devices
Table 2 Relative and overall UFP removal efficiencies for all air filtration devices tested.
Baseline conditions are also included
REMOVAL EFFICIENCIES FOR UFP
Panel Filters and Register Systems
Manufacturer
Air Cleaners, Inc.
Air Cleaners, Inc.
Camil Farr
Freudenberg Filtration Technologies L.P.
Freudenberg Filtration Technologies L.P.
eSpin Technologies, Inc.
eSpin Technologies, Inc.
eSpin Technologies, Inc.
IQ Air
IQ Air
Environmental Dynamics Group, Inc.

ID
BASELINE
WE1-7
WE2-7
CF-8
U-11
M-12
E-11
E-13
E-14
IQ-13
IQ-16

Relative Removal Efficiency
54
43
45
49
63
77
55
65
74
75
92

Overall Removal Efficiency
43
37
41
43
36
72
50
58
69
69
89

EDG

93

77

Stand-Alone Systems
Manufacturer
IQ Air
NQ Industries Inc.
NQ Industries Inc.
Pure Air

ID
BASELINE
CZ
NQ-1
NQ-2
PA

Relative Removal Efficiency
13
98
100
100
94

Overall Removal Efficiency
*
*
*
*
*

Table 3 Relative and overall BC removal efficiencies for all air filtration devices tested. Baseline
conditions are also included
REMOVAL EFFICIENCIES FOR BC
Panel Filters and Register Systems
Manufacturer
Air Cleaners, Inc.
Air Cleaners, Inc.
Camil Farr
Freudenberg Filtration Technologies L.P.
Freudenberg Filtration Technologies L.P.
eSpin Technologies, Inc.
eSpin Technologies, Inc.
eSpin Technologies, Inc.
IQ Air
IQ Air
Environmental Dynamics Group, Inc.

ID
BASELINE
WE1-7
WE2-7
CF-8
U-11
M-12
E-11
E-13
E-14
IQ-13
IQ-16

Relative Removal Efficiency
30
26
16
26
61
63
31
39
63
74
91

Overall Removal Efficiency
20
18
12
13
51
54
26
34
60
64
89

EDG

92

52

ID
BASELINE
CZ
NQ-1
NQ-2
PA

Relative Removal Efficiency
23
90
93
94
83

Stand-Alone Systems
Manufacturer
IQ Air
NQ Industries Inc.
NQ Industries Inc.
Pure Air

19

Overall Removal Efficiency
*
*
*
*
*

 Draft Report: Performance Evaluation of Air Filtration Devices
As expected, our data indicate that relative and overall removal efficiencies for all panel
filters increase with increasing MERV rating. In particular, filters WE1-7 and WE2-7 (both rated
as MERV 7) demonstrated the worst performance under this particular experimental set-up, with
removal efficiencies varying from 37 to 45% for UFP and from 12 to 26% for BC (Tables 2 and
3). These values are similar to or lower than those observed during our baseline measurements
and show no significant improvement in indoor air quality with respect to pre-testing conditions
with a standard filter. Conversely, filter IQ-16 (MERV 16) was able to remove a more
substantial portion of UFP and BC from the test classroom, and its overall removal efficiency
was 89% for both UFP and BC. In this respect, the IQ-16 filter was the only panel filter that met
the desired specifications set by AQMD and previously discussed under the Performance
Evaluation section.
Over the short time span of these measurements there was no significant increase in the
pressure drop across the panel filters tested (Table 1). The only register-system tested during this
study (EDG) was characterized by overall removal efficiencies lower than the guideline value set
by AQMD for this type of device, although the relative removal efficiency was higher than 85%
for both UFP and BC. A potential explanation for this discrepancy may be unfiltered air flow
escaping the register system in a way that would not affect measurements at position #4 but
would affect measurements at position #2.
It is worth noting that when testing stand-alone units the concentration of UFP (or BC)
inside the test classroom (position #2) was affected more by the classroom configuration, the
proximity of the stand-alone device to the air supply registers, the relatively high AER (2.1 hr-1),
the relative distance between the indoor real-time monitors and the air supply registers, the lower
air flow rate “processed” by these air filtration devices (between 350 and 545 cfm) relative to the
flow rate handled by the HVAC system, and other factors intrinsic to this particular classroom
set-up. The UFP and BC data obtained at positions #2 and #3 (Figure 2) and, thus, the overall
removal efficiency of the stand-alone units were not considered in the determination of their
performance. The outlet concentration (position #4) was used instead to evaluate the filtration
ability of the unit itself (i.e. relative removal efficiency). In this respect, all stand-alone devices
tested showed high relative removal efficiencies with values varying from 94 to 100% and from
83 to 94% for UFP and BC, respectively.
Sound Level Measurements
As noted earlier, the sound level from the stand-alone units was measured at 6 different
locations inside the test classroom (Figure 3). These readings were made because many school
districts have established a noise level below 45 decibels [db(A)] for new in classroom
equipment. Table 4 summarizes the noise level measurements made before and after turning the
stand-alone units on. The only noise heard with the stand-alone units off was the clicking from a
clock on the wall near the entrance door. None of the CPCs or Aethalometers were running at the
time of these noise-level tests. According to our results only the IQ CleanZone SL meets the 45
db(A) noise requirement set by AQMD.

20

 Draft Report: Performance Evaluation of Air Filtration Devices
Table 4 Noise level measurements for all stand-alone units
NOISE LEVEL dB(A)
Location
1*
2*
3*
4**
5***
6****

Distance
from source
40"
10' 40"
20' 40"
20' 40"
20' 40"
30' 40"
Average

IQ CleanZone SL
Off
38.2
35.1
44.2
39.2
39.8
43.9

On
44.1
43.0
48.6
41.2
42.2
47.0
44.4

Pure Air HPS350
Off
31.8
30.5
48.3
38.5
38.3
45.7

On
63.1
57.1
55.6
53.4
53.6
53.2
56.0

NQ400
Off
33.0
30.6
42.3
42.0
40.0
44.8

On
61.7
58.6
56.0
54.4
54.7
53.7
56.5

NQ Clarifier
Off
30.0
32.2
41.5
39.8
36.5
47.0

On
56.6
53.5
51.0
50.6
49.5
48.5
51.6

* In direct line with stand alone unit
** ~5' to left of direct line, but noise monitor was pointed toward stand alone.
*** ~7' to right of direct line, but noise monitor pointed toward stand alone.
**** Within 7' of rear wall

Ozone Measurements
Both stand-alone units manufactured by NQ Industries (NQ-1 and NQ-2) have one or
more ultraviolet lamps to remove germs and bacteria from the filtered air. Also, the register
system by Environmental Dynamics Group, Inc. (EDG) is supplied with an electrostatic filter to
enhance its performance. Therefore, these devices could potentially generate small but non
negligible amounts of ozone, an air pollutant that may worsen chronic respiratory diseases such
as asthma. Although we planned to monitor ozone upwind and downwind of these systems while
they were in operation, because of a temporary malfunction of the ozone monitors this part of the
testing protocol took place at the CE-CERT’s lab. The results of our measurements indicated that
these units do not generate any ozone.

21

 Draft Report: Performance Evaluation of Air Filtration Devices

REFERENCES
Abt, E., Suh, H.H., Catalano, P.J., Koutrakis, P. (2000) “Relative Contribution of Outdoor and
Indoor Particle Sources to Indoor Concentrations”, Environmental Science &
Technology, 34:3579-3587
Pope, C. A. and Dockery D. W. (2006), “Health effects of fine particulate air pollution: Lines
that connect”, Journal of the Air & Waste Management Association, 56:709-742
Environmental Protection Agency Integrated Science Assessments (External Review Draft;
2009): http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=201805
Shendell, D.G, Winer, A.M., Weker, R. and Colome, S.D. (2004) “Evidence of inadequate
ventilation in portable classrooms: results of a pilot study in Los Angeles County”,
Indoor Air, 2004; 14: 154–158
South Coast Air Quality Management District & IQAir North America, (2009) “Pilot Study of
High Performance Air Filtration for Classrooms Applications”

22

 Draft Report: Performance Evaluation of Air Filtration Devices

APPENDIX

Figure A-1 Front of portable classrooms at Sunnyslope elementary. All tests were conducted
inside the center unit

Figure A-2 Back of portable classrooms at Sunnyslope elementary. All tests were conducted
inside the center unit

Figure A-3 Bard WH483-A10xx4xxx heat pump

23

 Draft Report: Performance Evaluation of Air Filtration Devices

Figure A-4 CPC and Aethalometers used for measuring UFP and BC in ambient air and near the
air return duct (positions #4 and #3, respectively)

Figure A-5 CPC and Aethalometer used for measuring UFP and BC inside the classroom and at
breathing level (position #2)

Figure A-6 Left: Inlet side of the existing filter installed on 4/22/10, months before the
beginning of our tests. Right: unused filter (of the same type and brand as the one on the left)
used for the baseline measurements; installed on 7/7/10

24

 Draft Report: Performance Evaluation of Air Filtration Devices

Figure A-7 Left: Outlet side of the existing filter installed on 4/22/10. Right: unused filter (of the
same type and brand as the one on the left) used for the baseline measurements; installed on
7/7/10

Figure A-8 Inlet side of the filter installed on 7/7/10 for the baseline measurements; removed on
7/14/10

25