United States
Environmental Protection
Agency
National Risk Management
Research Laboratory
Research Triangle Park, NC 27711
Research and Development
EPA/600/SR-97/064 August 1997
v°/EPA Project Summary
Radon Diagnostic Measurement
Guidance for Large Buildings
Marc Y. Menetrez and Russell N. Kulp
The purpose of this study was to
develop radon diagnostic procedures
and mitigation strategies applicable to
a variety of large non-residential build-
ings commonly found in Florida. The
investigations document and evaluate
the nature of radon occurrence and en-
try mechanisms for radon, the effects
of heating, ventilation, and air-condi-
tioning (HVAC) system configuration
and operation on radon entry and dilu-
tion, and the significance of occupancy
patterns, building height, and other
building construction features. A pri-
mary focus of this project was the ef-
fect of the HVAC systems of a large
building on the transport, entry, and
hopefully the minimization of indoor
radon in the building. Two buildings
were investigated, both of which
showed an inverse relationship between
dedicated ventilation air and indoor ra-
don concentrations, as was expected.
Both also showed signs of unusual
HVAC design, operation, and mainte-
nance that presumably adversely af-
fected indoor radon and other indoor
air quality (IAQ) variables. The second
building showed clear indications of
foundation design elements that con-
tributed to radon entry. Some recom-
mendations relevant to building stan-
dards can be concluded from this
project. First, design and construction
should concentrate on elimination of
major soil gas pathways such as hol-
low walls and unsealed utility penetra-
tions. Second, HVAC system design
should include strategies designed to
minimize depressurized zones adjacent
to the soil. Third, while increased sup-
ply ventilation is generally helpful for
radon control, it is clearly not the most
cost-effective solution or prevention
tool once the requirements of occu-
pant comfort and general IAQ have
been met.
This Project Summary was developed
by EPA's National Risk Management
Research Laboratory's Air Pollution
Prevention and Control Division, Re-
search Triangle Park, NC, to announce
key findings of the research project
that is fully documented in a separate
report of the same title (see Project
Report ordering information at back).
Introduction
This report describes the results of a
project conducted by Southern Research
Institute and other organizations for the
U.S. Environmental Protection Agency on
behalf of the Florida Department of Com-
munity Affairs. The purpose of this study
is to develop radon diagnostic procedures
and mitigation strategies applicable to a
variety of large non-residential buildings
commonly found in Florida. To accom-
plish this, it was necessary to perform
detailed field investigations and paramet-
ric studies in a variety of buildings that
have elevated levels of radon. The inves-
tigations document and evaluate the na-
ture of radon occurrence and entry mecha-
nisms for radon, the effects of HVAC con-
figuration and operation on radon entry
and dilution, and the significance of occu-
pancy patterns, building height, and other
building construction features.
A primary focus of this project was the
effect of the HVAC systems of a large
building on the transport, entry, and hope-
fully the minimization of indoor radon in
the building. The full report discusses
HVAC systems and their effects and de-
scribes case studies in two large buildings
-------
in Florida. Conclusions and recommenda-
tions address elements of significance to
proposed statewide standards for radon
resistance in new large building construc-
tion.
Case Study 1: Financial Center
North
The first large building selected for a
radon case study is the Financial Center
North (FCN) building in Deerfield Beach,
Florida. This is a privately owned building
that is being leased to the General Ser-
vices Administration (GSA) for purposes
of housing FCN of the Internal Revenue
Service (IRS). Crown Diversified Indus-
tries Corporation (CDIC) owns the build-
ing. The building is a combination office
and warehouse/maintenance facility. It is
constructed in two wings in the shape of
an L. Each wing has three floors: each
floor in the north wing is about 5600 ft2*
(112 x 50 ft), and each east wing floor is
about 6200 ft2 (124 x 50 ft). The ware-
house/maintenance portion of the facility
is in the crook of the L. It is predominantly
a two-story high-bay space. This area of
the building is used primarily by a mainte-
nance staff that services an adjacent apart-
ment complex also owned by CDIC. The
maintenance/warehouse area is about
10,450 ft2 (110 x 95 ft). The entire building
is about 46,000 ft2 and can accommodate
about 125 occupants.
The HVAC systems are of theUnitary
type and rely on 22 separate direct ex-
pansion split systems for primary cooling
to the office spaces. All of the condensers
are frame mounted and located on the
roof. The system evaporators are located
in ceiling-hung air handlers (AHs) units in/
or near the comfort zone being served. In
addition to housing the evaporator, all of
the AHs contain electric reheat coils that
provide space heating. The AHs are also
provided with a system of distribution
ductwork consisting of supply, return, and
outdoor air (OA) connections. Cooling and
heating of the occupied space is con-
trolled by wall mounted thermostats. Each
AH has its own individual thermostat. The
heating and cooling capacities of each
split system range in size from 2.91 tons
cooling/7.2 kW heating to 9.16 tons cool-
ing/15 kW heating.
Air is exhausted from the building pri-
marily by three roof-mounted power roof
* For readers more familiarwith metric units: 1ft2=929
cm2,1 ft = 0.305 m, 1 ton = 907 kg (or 0.907 metric ton),
1°F = 9/5°C + 32,1 in. = 2.54 cm, 1 in. WG = 249 Pa,
1 cfm = 0.0283 nf/min, and 1 pCi/L = 37 Bq/m3.
ventilators (PRVs), as well as toilet ex-
haust fans. The original HVAC design
called for the OA to be provided through
two outdoor air risers (OARs) that, through
a system of ductwork, were connected to
the suction side of each AH. None of the
OARs were originally powered by a fan.
The introduction of OA was reliant on the
ability of the AH fan to inject OA from the
roof level down the OAR and into the
intake of the AH. The original design speci-
fied that a total of 4500 cfm of OA be
introduced to the building. This quantity of
OA represents 10% of the total building
supply air.
The FCN Building has initially exhibited
radon levels of approximately 10 picocuries
per liter (pCi/L), during GSA screening
measurements, which are above the EPA
action level guideline of 4 pCi/L. In early
1992, Radon Environmental Testing Cor-
poration was requested to provide radon
measurement and mitigation service to the
building management. Passive sealing of
slab cracks and penetrations was provided
as well as increasing the level of OA by
installing supply fans. This reduced radon
levels down below the 4 pCi/L guideline
and generally subjectively improved IAQ.
New OA fans raised the level of OA
from 21 to about 66% of design. Using
ASHRAE Standard 62-1989, this new
quantity of OA would support 200 occu-
pants (3000 cfm at 15 cfm per occupant).
Past reports indicate not more than 102 oc-
cupants in the building at any time (17
occupants per floor x 6 floors). At the time
of our study, the building was being oper-
ated in this mode.
Experimental Plan: Outdoor Air
Variations
For this part of the study, it was agreed
that the primary feature of the HVAC sys-
tems in mitigating radon is pressurization
of the building.
It was decided to operate the HVAC
systems in four different modes of build-
ing pressurization while collecting data.
These modes of operation were deter-
mined by our ability to vary and control
the amount of OA allowed to be intro-
duced into the building while maintaining
supply and exhaust at known quantities.
The four modes consisted of operation of
the system(s) to provide 0, 5, 10, 15, and
20 cfm/occupant from the OA supply fans.
No changes in the supply or exhaust air
quantities were made. These predeter-
mined modes of operation describe situa-
tions from complete system shutdown of
OA quantities to those recommended in
ASHRAE Standard 62-1989. Model, no
OA, would be considered the worst case
scenario. Under this mode of operation,
the building is under complete negative
pressure and all OA enters by infiltration.
As OA supplied to the AHs increases,
infiltration decreases, resulting in no
change in supply or exhaust quantities,
although increasing OA causes increased
pressurization throughout the building.
Mode 2 would simulate the OA require-
ments illustrated in the Florida Adminis-
trative Code (FAC) chapter 6A-2 that con-
trols the amount of OA to 5 cfm/occupant.
Modes 3 and 4 would be variations on the
ASHRAE Standard 62-1989, using 15 and
20 cfm/occupant.
For Mode 1, the OA intakes were closed
with polyethylene to ensure a complete
nonporous seal. Mode 1 was accomplished
over a weekend (July 3-6, 1992) since the
building owner would not permit the HVAC
systems to be operated without OA during
normal working hours. For Mode 2 (July
6-15), the HVAC systems were balanced
so that the measured OA intake was actu-
ally 5.5 cfm/occupant. Mode 3 (June 16 -
July 3) was measured at 13.6 cfm/occu-
pant. Mode 4 (July 15-27) was 19.5 cfm/
occupant.
Data were collected from the data sta-
tions by downloading data files through
the internal modem by telephone connec-
tion. The information was converted into
usable numbers, calibrated, and put into
graphs and tables. Data files were ana-
lyzed and compared with other informa-
tion, such as maintenance practices. The
FCN data are limited in scope due to
instrumentation difficulties that were cor-
rected for the second case study. FCN
results are limited to radon concentrations
and some perfluorocarbon tracer (PFT)
gas measurements.
By intentionally reducing the OA intake,
an increase in radon concentrations was
exhibited to a peak level above 4 pCi/L
throughout the building. Distinct average
levels of radon can be identified from the
data for a consistent level of OA intake. A
comparison of radon levels versus OA
intake flowrate is evident in the averaged
continuous radon monitor data. The build-
ing average concentration at 0 cfm (per
occupant) was 2.6 pCi/L as compared to
1.8 pCi/L at 5.5 cfm, 1.2pCi/L at 13.6
cfm, and 1.0 pCi/L at 19 cfm. A reduction
correlated with increased OA is clearly
present; however, due to imprecision in
measurement and expected fluctuations
in radon concentrations, it is not possible
to form quantitative conclusions.
Case Study 2: Polk County Life
and Learning Center
The second case study in this project
was conducted at the Polk County Life
and Learning Center (LLC). While some
-------
of the same measurement techniques were
used at this building as for the previous
study at the FCN, the experimental and
analytical sequences were much more
detailed. The LLC consists of three build-
ings: The Center for the Trainable Men-
tally Handicapped, the Severely Handi-
capped Center (two classroom addition),
and the Greenhouse. The Center for the
Trainable Mentally Handicapped and the
Greenhouse were designed in 1974 and
built in 1975. The Severely Handicapped
Center was designed in 1984 and con-
structed in 1985. This study focuses en-
tirely on the LLC for the Trainable Men-
tally Handicapped. The LLC building is a
single-story training/school building of
about 18,000 ft2. It consists of staff office
space, classrooms, a large multipurpose
room, a kitchen area, janitorial closets,
and a woodshop. The facility houses 103
students daily along with 22 staff mem-
bers for a total of 125 occupants. Archi-
tecturally, the building is constructed as a
slab-on-grade. The slab is 4 in. reinforced
concrete on compressed fill and provided
with a vapor barrier. The vapor barrier is
assumed to be polyethylene (the draw-
ings are not specific). The walls of the
Center are 8 in. CMUs (concrete masonry
units; i.e., blocks) with stucco exterior and
5/8-in. gypsum boards on 1 x 2 in. furring
strips on the interior. The roof system
consists of wood truss construction with
asphalt shingle roof tiles over most of the
roof. However, in some areas the roofing
consists of rolled mineral roofing material.
The interior ceilings are either lay-in tile or
painted gypsum board. All windows are
either aluminum frame single hung or bay
windows. The interior walls are gypsum
board on wood studs with interior ceiling
heights of typically 9 ft except for the
central area used as a cafeteria/audito-
rium. The LLC is divided into four fire
control zones by means of rated 5/8-in.
gypsum board that extends to the tectum
decking below the roof. However, many
openings between zones (some as large
as 2 x 4 ft) tend to merge the separate
zones into one or two larger zones. Fire
walls divide the building into four zones
(although the zones appear to be well
coupled).
The LLC is heated and cooled by an
ALL-AIR system composed of a single
main AH. The AH provides cooling to the
LLC by means of a 21 ton (252 x 106 Btu/
hr) direct expansion split system and a
distribution system of supply ductwork. The
system is low pressure (2.5 in. WG) and
uses a single supply duct and a ceiling
plenum return air system. The individual
rooms and zones are environmentally con-
trolled by variable-air-volume (VAV) boxes
mounted above the ceiling in the return
plenum. Wall-mounted thermostats con-
trol the VAV boxes. The introduction of
OA is controlled by a roof-mounted supply
fan. This fan initially could provide up to
1200 cfm of unconditioned OA directly to
the AH return air plenum. The OA then
mixes with the building return air. The AH
has the capacity to supply 5620 cfm at
conditions of 57°F dry bulb/56°F wet bulb
which gives the machine a rating of ap-
proximately 21 tons. This is approximately
1.2 tons per 1000ft2 which constitutes a
greatly oversized capacity. Based on the
Florida Administrative Code, Chapter 6A-
2 requirements of 5 cfm/occupant, the LLC
could house 240 occupants. The actual
occupancy of 120 people increases the
OA to 10 cfm/occupant (1200 cfm/120 oc-
cupants). Heat for the LLC building (via
AH) is provided by a 15-kW strip heater.
In addition, each VAV box that serves a
space that is adjacent to an exterior wall
is provided with an additional strip heater.
VAV box strip heaters are controlled by
room-wall-mounted thermostats. The build-
ing is served by 26 VAV boxes that are
sized for a full air-conditioning load of
11,305 cfm. The boxes are set for a mini-
mum setting of 40% of full load. The di-
versity factor (100 times the ratio of the
sum of the individual VAV box capacities
divided by the AH capacity; i.e., 100 x
[5620/11,305]) is calculated to be approxi-
mately 50% for the VAV box operation.
Exhaust air from the LLC is through 14
exhaust fans located in the toilets, bath-
rooms, janitor closets, workshop, and
kitchen. The total design building exhaust
from these 14 fans is 2350 cfm when all
are operating.
Initial radon measurements were made
at the LLC by the Polk County Health Unit
during the 1990-91 and 1991-92 school
years. The radon levels averaged 10.4
pCi/L and were fairly independent of the
seasons.
Based upon inspection of the design
plans for the building, it was easy to see
that this building may have been operated
in an undesirable HVAC negative pres-
sure mode. Since the maximum OA quan-
tity was 1200 cfm and the exhaust quan-
tity is 2350 cfm, the building may have
been operated negative by about 1150
cfm or less. To compound this imbalance,
the OA fan was set to shut off when the
return air temperature was below 70°F or
above 80°F. The fan controls would only
allow the OA fan to operate when the
return air temperature was in the range of
70-80°F. Further, the OAfan was installed
backwards on the motor shaft, and the
motorized damper for the OA fan was
frozen in the closed position. Other defi-
ciencies identified include: leakage from
the supply air on both sides of the VAV
boxes, and in the main supply duct feed-
ing all the VAV boxes, several VAV con-
trol mechanisms inoperative, and four ex-
haust fans inoperative. A list of the build-
ing deficiencies was sent to school offi-
cials on about November 10, 1992. It was
agreed that the Polk County School Sys-
tem would fund all "punch list" items and
the EPA would fund the test and balance
(TAB) fee and fan replacement.
The LLC was instrumented with five of
the EPA Data logging systems on Octo-
ber 27-29, 1992. School maintenance per-
sonnel implemented a repair procedure at
the LLC to correct the deficiencies de-
tected in the building during the walk-
through on November 5, 1992, and as
described in the building pre-balance sur-
vey carried out by the TAB company.
These repairs were completed during the
latter part of January 1993. During De-
cember, the Phoenix Agency, Inc. (PAI)
replaced the OA supply fan and damper.
The as-found condition of the building was
such that little or no OA was being sup-
plied to the building. The only ventilation
was through openings in the building shell.
This was evident from the odors that per-
sisted in several rooms; in particular, in
Room 105. The new OA fan can supply
3000 cfm of OA.
1993 Parametric Study
Testing at the LLC was carried out us-
ing the following conditions. OA flowrates
of 0,750, 1500, 2250, and 3000 cfm (0, 5,
10, 15, and 20 cfm/person) were used.
Generally, each OA flowrate condition oc-
cupied a week of testing. The exhaust
fan-on condition was maintained over 1-1/
2 days of the weekend, and the exhaust
fan-off condition was maintained at night
and the remainder of the weekend. Typi-
cally, the HVAC fan operated on a 12
hour on/12 hour off cycle each day.
The radon levels in the LLC building
were significantly reduced from the levels
first measured in December 1992. Aver-
aged radon levels were measured in
Rooms 102, 109, the Cafeteria, the Audi-
ology room, and the Conference room with
the Femto-Tech continuous monitors at-
tached to the EPA data loggers. Several
aspects of the data are apparent. First,
the levels measured during December
1992 and January 1993 were much higher
than those measured by the Polk County
Health Unit. The reasons for this large
difference are not known. Second, the
overall levels show a steady decrease as
shown by the 5-day moving (un-weighted)
average line. This is due primarily to the
replacement of the OA fan and damper,
-------
and to the consistent operation of this fan.
Also, the new OA fan greatly reduced the
level of offensive odors noticed in Room
105 in October 1992. The second inter-
esting aspect of the data is the fine struc-
ture or daily variations in the radon levels.
These are caused primarily by the daily
cycling of the HVAC system from daytime
use to nighttime setback. Several aspects
of the radon time series data were readily
apparent:
1. The radon levels generally increase
overnight until the HVAC system
comes on.
2.
3.
Once the HVAC system turns on,
the levels drop rapidly.
As the HVAC system operates, the
levels drop but seldom go below 4-
5 pCi/L. The rate of drop and the
limiting radon level depend as ex-
pected on the OA flowrate.
4. When the exhaust fans are run con-
tinuously, the radon levels do not
increase to nearly as high levels
with the HVAC system off as they
do when the fans are left off.
5. The radon levels in the Audiology
room do not always follow the rest
of the building. The levels here are
usually higher than the building as
a whole.
From these observations several con-
clusions appear obvious. The HVAC sys-
tem is assisting in lowering the radon lev-
els even without the intentional introduc-
tion of OA. A significant factor is almost
certainly the enhanced ventilation rate in-
duced by the system. Pressure differences
across the building shell will enhance infil-
tration through shell openings, especially
when exhaust fans are active. This infil-
trating air is difficult to measure and
changes the definition of "no OA" to mean,
"no OA actively supplied by the OA sup-
ply fan." The peak radon levels reached
in the building just before the HVAC sys-
tem comes on do depend somewhat on
the amount of OA introduced into the sys-
tem during the previous HVAC operation
cycle and on the length of time that has
transpired since the HVAC system was
last operated. The main reasons for the
persistently higher radon levels in the Au-
diology room were thought to be due to
the isolation of the room combined with a
major entry path such as an open (or
extremely leaky) expansion joint located
under the room.
In an effort to understand why the ra-
don levels remain at 4 pCi/L or greater,
two additional Pylon AB5 continuous ra-
don monitors with PRD-1 passive cells
were placed outside the building in the
sheltered workshop area on the north side
of the building. One monitor was located
at ground level and the other about 8 ft off
the ground. The locations were open and
sheltered only from rain. Over the 4-week
period the ground level radon averaged
4.1 pCi/L with a weekly high of 5.3 pCi/L
(4/30-5/7) and a weekly low of 2.8 pCi/L
(4/16-4/23). These results were of obvi-
ous concern since, if the radon source
strength is sufficiently high, the indoor lev-
els can never be reduced below the aver-
age ground level outdoor levels.
1994 Phase II Study
In order to address some of the uncer-
tainties in the 1993 results, permission
was obtained for a series of follow-up
tests at the LLC. Continuous measure-
ments by Southern Research Institute rep-
licated some of the conditions studied in
1993. Significant changes included:
• Three outdoor air radon monitors were
installed to investigate the distribution
and time variability of outdoor radon.
• Since the indoor radon was known to
be well-mixed outside the Audiology
room, only one indoor radon monitor
(in the Cafeteria) was used. An addi-
tional monitor was installed in the Me-
chanical room.
• In order to investigate the significance
of the load-bearing block walls as an
entry route, a pumped radon monitor
was used to sample the air within
one section of the block wall cavity.
Another pumped radon monitor was
used to sample subslab radon con-
centrations.
• Pressure differentials (with respect to
the Cafeteria) were monitored in the
following zones: the Mechanical room,
outdoors, subslab, and the block wall
section.
• Sulfur hexafluoride (SFej was continu-
ously injected into the Cafeteria (and
generally found to be uniformly dis-
tributed in the building).
• In addition to operation of the HVAC
during several week-long periods at
each of the OA damper positions used
previously, several periods of depres-
surization (using the exhaust fans)
were scheduled on weekends. One
period of pressurization (with the
HVAC off) was performed using a
blower door. Out of deference to the
energy management concerns of the
school district, the customary setback
schedule (8 hr on/16 hr off, 5 days/
week) was used in the 1994 study.
The outdoor radon experiments indicate
that outdoor radon was not a significant
source of the indoor radon at the LLC.
The monitor sampling at ground level (3
in.) failed early in the study and was re-
moved. The other monitors (at 4 ft and on
the roof level at the OA intake) continued
to operate throughout the study period.
Both monitors showed low values during
the day (typically <0.5 pCi/L) followed by
peaks of 2-6 pCi/L or higher at night,
when turbulent mixing is low. The OA
contribution is seen to be minimal, since
the OA concentration is at background
levels during the day when the building
ventilation rate is significant. In the early
morning hours when the outdoor radon
concentration is highest, the indoor con-
centrations are several times higher; fur-
thermore, the infiltration rate is quite low
at these times.
Examination of differential pressure data
suggests further insights into the normal
operating state of the building. These data
cover week-long periods of normal opera-
tion with the OA damper set at positions
corresponding to nominal flowrates of
3000, 750, 2250, and 1250. First, the Caf-
eteria runs at positive pressure with re-
spect to outdoors when the AH is operat-
ing. The mean pressurization varies from
about 0.3 Pa at 750 cfm nominal OA to
about 1 Pa at 3000 cfm nominal OA. Dur-
ing weekdays, this pressure differential
undergoes dramatic fluctuations. These
may be partly due to changes in the build-
ing load (VAV operation), or in the OA fan
operation, as they are not present during
periods with the HVAC off, even when the
building is mechanically pressurized or
depressurized. However, since these fluc-
tuations are also characteristic of occu-
pied periods, they may result from occu-
pant activity (i.e., opening doors or win-
dows). During HVAC off periods, the Caf-
eteria-outdoor pressure drops to low val-
ues, and a slight depressurization is ob-
served on many nights. This depressur-
ization is most likely explained by the ob-
servation that a few exhaust fans were
often left in operation after the staff left at
the end of the day. These unmonitored
changes in building operation were unfor-
tunate, since they leave some uncertainty
as to the exact operating mode of the
building.
The Mechanical room is depressurized
relative to the Cafeteria by 1.5-2.0 Pa when
the AH is operating. This difference is
greater than the pressurization of the Caf-
eteria, so the Mechanical room is nega-
tive with respect to outdoors for all but
brief portions of the normally occupied
periods. Pressures in the other zones
monitored (subslab, block wall, and Me-
chanical room during periods without AH
operation) track the Cafeteria pressure,
-------
but tend to be slightly lower in magnitude
during mechanical pressurization or de-
pressurization.
Examination of continuous radon data
reveals several trends with implications
for radon entry in the building. First, the
radon concentrations in the Mechanical
room are consistently higher than in the
Cafeteria under all HVAC and mechanical
ventilation conditions, although the differ-
ences grow smaller for periods of me-
chanical depressurization or of low OA
damper setting. Since ventilation rates be-
tween the two zones are expected to be
high, and inleakage of OA is expected to
be much higher into the Mechanical room
as compared to the Cafeteria, the higher
concentration suggests a significantly
higher radon entry rate into the Mechani-
cal room. This is not surprising in light of
the pressure measurements showing the
Mechanical room to be the most highly
depressurized portion of the building.
A second observation is that the ratio of
indoor radon to SF6 tracer is significantly
higher in the daytime (with HVAC in op-
eration) and during depressurization, indi-
cating higher radon entry rates during
those periods. (Since the ventilation rate
is also increased, these periods tend to
be periods of lower radon concentrations.)
The increased entry rate when the AH is
turned on also helps explain the slower
rate of fall of radon concentration than
would be expected from the air change
rate. Indeed, the SF6 tracer gas does drop
much more rapidly to its limiting daytime
value. The slower decay time of the radon
is partially explained by the larger mea-
surement time constant of the radon moni-
tors due to decay times of the radon prog-
eny, but is also an indication of the change
in entry rate as the HVAC cycles between
normal and setback operation.
The block wall radon concentrations sug-
gest this to be a major pathway for radon
entry. During periods of AH operation, the
block wall radon rapidly rises to levels of
600-1000 pCi/L, then drops back to lower
levels during the evening setback period.
Inspection of the block wall pressure at
the section tested reveals that, while it is
generally positive with respect to outdoors
during HVAC on periods, it is generally
negative with respect to both the subslab
test point and the Cafeteria. This depres-
surization may be due to coupling with the
plenum, and would suggest one path for
transport of radon into the return air sys-
tem. This pressure coupling would explain
the rapid influx of soil gas into the block
wall cores as the AH starts in the morn-
ing, as is clearly seen in the continuous
data. The rapid drop in block wall radon
as the AH goes off in the afternoon can
be attributed to the relief of this driving
pressure gradient combined with trans-
port of the accumulated radon back into
the soil or, more probably, into the build-
ing. The likelihood that the block walls
provide a major entry path has been dis-
cussed before, since the cores of these
walls penetrate the slab to the block
courses in direct contact with the soil.
Since the pores of the block are highly
permeable, a low resistance pathway ex-
ists directly from the soil to indoors.
Two cautions must be observed regard-
ing any quantitative interpretations from
these results, however. The block wall
section tested was a 4 ft wide interior wall
segment bordering the Cafeteria and
Room 109. Horizontal communication
within the wall segment will presumably
be limited by the reinforced filled core
sections specified every 4 ft in this build-
ing. There are exhaust fans in Room 109,
which may enhance entry in this section
over walls adjacent to rooms with no me-
chanical exhausts. On the other hand, en-
try into the walls of the Mechanical room
might easily be much higher, and could
represent the major source of radon entry
into the building. In any event, the results
of the present study clearly indicate that
the wall construction detail used in the
LLC is highly vulnerable to radon entry,
and alternatives must be provided for a
radon-resistant building standard.
Conclusions and
Recommendations
The two case studies in this report
present some insight which can be gener-
alized to other structures. The first build-
ing was a structure that had apparently
been successfully mitigated by passive
techniques, so would not normally be con-
sidered a "problem" structure. In this build-
ing, variations in OA flow control dampers
produced ventilation rate changes within
the typical range (0.2 to 0.6 of an air
change per hour) resulting in variations in
indoor radon concentrations over a com-
parable range (a factor of 2.6). The sec-
ond building had much higher radon lev-
els, which could not be reduced below the
4 pCi/L radon standard without introduc-
ing OA at a rate in excess of ASHRAE
standard requirements, not to mention the
energy management priorities of the
owner. Both buildings demonstrated an
inverse relationship between dedicated
ventilation air and indoor radon concen-
trations, as was expected. Both also
showed signs of unusual HVAC design,
operation and maintenance which presum-
ably adversely affected indoor radon as
well as other IAQ variables. The second
building showed clear indications of foun-
dation design elements which contributed
to radon entry; elimination of these entry
paths at the time of construction would
have been by far the most cost-effective
remedy for the building.
Some recommendations relevant to
building standards can be concluded from
this project. First, design and construction
should concentrate on elimination of ma-
jor soil gas pathways such as hollow walls
and unsealed utility penetrations. It is clear
from this study how much benefit can be
derived from sealing of minor cracks and
joints. Second, HVAC system design
should include strategies designed to mini-
mize depressurized zones adjacent to the
soil. Such zones could be caused by flow
imbalance in the air distribution system,
inadequate sealing of major duct leaks, or
imbalance of supply and exhaust ventila-
tion airflow. The combination of depres-
surized areas and poor barriers is particu-
larly undesirable, especially if the depres-
surizing element is the return air portion
of the AH system. Third, while increased
supply ventilation is generally helpful for
radon control, it is clearly not the most
cost-effective solution or prevention tool
once the requirements of occupant com-
fort and general IAQ have been met.
The information base needs to be ex-
tended. In particular, monitoring of the ra-
don in new buildings constructed on high
radon potential soil according to radon
control guidelines could provide useful in-
formation.
-------
Marc Y. Menetrez (also the EPA Project Officer, see below) and Russell N. Kulp
are with EPA's National Risk Management Research Laboratory, Research
Triangle Park, NC 27711
The complete report, entitled "Radon Diagnostic Measurement Guidance for Large
Buildings, Volumes 1 and 2,)," (Order No. 600/R-97/046A [Volume 1] PB97-
189716 and 600/$-97/064B [Volume 2] PB97-189724; Cost: Volume 1 $31.00,
Volume 2 $35.00, subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Air Pollution Prevention and Control Division
National Risk Management Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
United States
Environmental Protection Agency
Center for Environmental Research Information
Cincinnati, OH 45268
Official Business
Penalty for Private Use $300
BULK RATE
POSTAGE & FEES PAID
EPA
PERMIT NO. G-35
EPA/600/SR-97/064
------- |